September 24, 2025 Dataset Restricted Access

Ratliff, Hunter; Blangiardi, Francesco; Kögler, Toni

JSON-LD (schema.org) Export

{
  "datePublished": "2025-09-24", 
  "@id": "https://doi.org/10.14278/rodare.3997", 
  "sameAs": [
    "https://www.hzdr.de/publications/Publ-43014"
  ], 
  "@type": "Dataset", 
  "keywords": [
    "proton therapy", 
    "treatment verification", 
    "particle transport calculations", 
    "PHITS"
  ], 
  "inLanguage": {
    "name": "English", 
    "alternateName": "eng", 
    "@type": "Language"
  }, 
  "version": "1.0.0", 
  "name": "PHITS simulations of neutron and gamma-ray production from and transport of 70\u2013250 MeV protons in heterogeneous 1D tissue phantoms", 
  "identifier": "https://doi.org/10.14278/rodare.3997", 
  "@context": "https://schema.org/", 
  "url": "https://rodare.hzdr.de/record/3997", 
  "description": "<p><strong>Introduction</strong></p>\n\n<p>This dataset corresponds to the PHITS simulation data used in &quot;Fast Phase Space Reconstruction for Proton Beam Traversal and Neutron Emission in Proton Therapy using Fourier Neural Operators&quot;.<br>\nA concise description of the simulation setup is provided here; please refer to the paper for detailed discussion, discription, analysis, and further results derived from this dataset.</p>\n\n<p><br>\n<strong>Description of simulations</strong></p>\n\n<p>This dataset consists of <a href=\"https://phits.jaea.go.jp/\">PHITS</a> simulations for 47 different proton energies from 70 MeV to 250 MeV incident upon different &quot;1D&quot; heterogeneous cylindrical phantoms (varied materials every 0.5 mm in length, uniform radially and rotationally) whose composition (materials and sequence along length) are taken from randomly sampled rays cast through a 3D CT phantom [CITE] with CT number mapped to material composition and density via the&nbsp;<em>HumanVoxelTable-KumamotoUniv.data</em>&nbsp;conversion table within the RT-PHITS utilitydistributed with PHITS.&nbsp;<br>\nIncluded tallies score spatial distributions of energy deposition, LET, proton current (with an additional angular dimension), neutron production, gamma-ray production, and a variety of diagnostic tallies.&nbsp;<br>\nEvent-by-event &quot;list-mode&quot; data is scored for neutron and gamma-ray production, called &quot;dump&quot; tallies in PHITS.</p>\n\n<p>Given the objective of these simulations was for AI model development, the 47 energies are divided into 37 <strong>training</strong> energies (70 MeV to 250 MeV in 5 MeV steps) and 10 <strong>testing</strong> energies (73 MeV to 245.8 MeV in 19.2 MeV steps).&nbsp;<br>\nFor each energy, two simulations were ran: (1) a simulation with <strong>1E8</strong> protons simulated where all <strong>tallies</strong> (including <strong>dump</strong> tallies) were included/enabled and (2) a simulation with <strong>1E9</strong> protons simulated (available on request) but with only <strong>dump</strong> tallies enabled (other tallies disabled to reduce memory consumption and increase simulation speed).&nbsp;<br>\nFurthermore, all of the above was actually performed twice: (1) initially with purely <strong>monoenergetic</strong> beam energies and with a spatial spread of 2.5 mm and (2) a second &quot;more realistic&quot; set with <strong>Gaussian-distributed</strong> energies (with energy-dependent FWHM) and slightly wider 4.0 mm beam spread.</p>\n\n<p>All simulation outputs were automatically processed from the plaintext and binary files produced by PHITS into compressed pickle file objects (NumPy arrays, Pandas DataFrames, dictionaries) using the <a href=\"https://github.com/Lindt8/PHITS-Tools\">PHITS Tools</a>&nbsp;</p>\n\n<p>Python utility.&nbsp;<br>\nThese Python objects were then utilized in the subsequent analysis of the paper this simulation set was generated for.</p>\n\n<p><br>\n<strong>Structure of this repository</strong></p>\n\n<p>The volume of data present in this repository is quite substantial (~ 700 GB).&nbsp;<br>\nTherefore, the repository has been structured in a way to allow flexibility in only downloading data of interest.</p>\n\n<p>The root directory of this repository consists of 39 top-level directories whose names indicate their contents.<br>\nWithin each are two directories:&nbsp;<em>training</em> and&nbsp;<em>testing</em>.<br>\nWithin each of these are directories of the format <em>???_MeV</em>, where <em>???</em> is replaced by three digits specifying the nominal beam energy in MeV.&nbsp;<br>\n(This is <em>???p?</em> for the energies of the testing dataset, with <em>p</em> in place of a decimal point.)<br>\nThus, each <em>training</em> directory contains 37 subdirectories, and each <em>testing</em> directory contains 10 subdirectories.<br>\n(One should note that there are no setup differences between <em>training</em> and <em>testing</em> data; they are simply divided here in the same way as in the paper.)<br>\nEach <em>???_MeV</em>/<em>???p?_MeV</em> directory contains simulation input/output and/or PHITS Tools processed output, depending on the top-level directory it is contained within.<br>\nInput and output file names do not differ between different energies; directory structure is used to keep them distinguished/separated.</p>\n\n<p><strong>PHITS input information</strong></p>\n\n<p>One top-level directory differs from all of the others, and this is common_inputs.<br>\nAs the name suggests, this directory contains all PHITS input information used in generating all of the simulation outputs.&nbsp;</p>\n\n<p>The core two PHITS input files used are <em>beam-on-target_phits-input_MonoE.inp</em> for the monoenergetic beam simulation set and <em>beam-on-target_phits-input_GaussE.inp</em> for the Gaussian-distributed beam energy simulation set.<br>\nWithin these inputs are lines using the PHITS insert file function <em>infl:{*}</em>; all inserted files used in the PHITS simulations are also contained within this <em>common_inputs</em> directory.<br>\nThe single exception to this is <em>PARAMETERS_files-1-and-7.txt</em>, which is simply the <em>file(1)</em> and <em>file(7)</em> PHITS <em>[Parameters]</em> arguments and will be system-specific paths to PHITS installation/data files.<br>\nAlso note that relative paths are used in the <em>infl:{*}</em> commands; these relative paths differ to how this repository is structured given the repository has been restructured in post for distribution convenience.&nbsp;<br>\nFile names are still unique and can be found in this <em>common_inputs</em> directory.<br>\nThe <em>CELL</em> subdirectory contains the <em>[Cell]</em> sections used for the varied phantom compositions, and the <em>MAPPINGS_OF_ENERGY_TO_CELL_FILES.csv</em> file details how these files are paired with the 47 different beam energies.</p>\n\n<p><strong>PHITS outputs (raw and processed)</strong></p>\n\n<p>The remaining 38 top-level directories contain simulation/processed output.<br>\nWhen these simulations were ran, all output was contained in each <em>???_MeV</em> directory.&nbsp;<br>\nAs detailed earlier, these have been split into various top-level directories here to allow more convenient download of only desired files.<br>\nNominally, each of these <em>???_MeV</em> directories contained the following before being split:</p>\n\n<ul>\n\t<li>a <em>beam-on-target_phits-input.inp</em> PHITS input file (and a simple <em>phits.in</em> pointing to this input file, needed for parallel running of PHITS); note that these inputs have all specific source energy information populated within this file</li>\n\t<li>a <em>phantom_composition_info.csv</em> file also detailing the phantom composition used for that beam energy</li>\n\t<li><em>tphits*.out</em> file(s),<strong>&nbsp;raw</strong>&nbsp;summary output files generated by PHITS</li>\n\t<li><em>*.out</em> <strong>raw</strong> plaintext tally output files from PHITS</li>\n\t<li><em>*.eps</em> <strong>graphical</strong> visualizations of tally output, generated by PHITS</li>\n\t<li><em>*_dmp.out*</em> <strong>raw</strong> binary tally <strong>dump</strong> files from PHITS</li>\n\t<li><em>*.pickle.xz</em> <strong>processed</strong> tally output (and <em>phits.out</em> metadata) from PHITS Tools, LZMA-compressed pickle files</li>\n\t<li><em>*_dmp_namedtuple_list.pickle.xz</em> <strong>processed</strong> tally <strong>dump</strong> output from PHITS Tools, formatted as a NumPy record array (np.recarray)</li>\n\t<li><em>*_dmp_Pandas_df.pickle.xz</em> <strong>processed</strong> tally <strong>dump</strong> output from PHITS Tools, formatted as a Pandas DataFrame (same numerical data as in NumPy recarray)</li>\n\t<li><em>*.png</em> and <em>*.pdf</em> <strong>graphical</strong> visualizations of tally output, generated by PHITS Tools</li>\n</ul>\n\n<p>The top-level directories of this repository are named in a way to detail (1) which simulations their contents pertain to and (2) which output files are contained within them.<br>\nThe directories are named using an underscore-delimited pattern whose components have the following names and meanings:</p>\n\n<ul>\n\t<li>Beam type:&nbsp;\n\t<ul>\n\t\t<li>&nbsp;<strong>MonoE</strong> refers to simulations with the monoenergetic beams with 2.5 mm spread</li>\n\t\t<li>&nbsp;<strong>GaussE</strong>&nbsp;refers to simulations with the Gaussian-distributed energies and 4.0 mm spread</li>\n\t</ul>\n\t</li>\n\t<li>Simulated number of protons:\n\t<ul>\n\t\t<li><strong>1E8</strong>&nbsp;refers to simulations with 10<sup>8</sup> (one hundred million) protons simulated&nbsp;</li>\n\t\t<li><strong>1E9</strong> refers to simulations with 10<sup>9</sup> (one billion) protons simulated (only available on request)</li>\n\t</ul>\n\t</li>\n\t<li>Output source/type:\n\t<ul>\n\t\t<li><strong>raw</strong> refers to the PHITS input and PHITS-generated output&nbsp;</li>\n\t\t<li><strong>processed</strong>&nbsp;refers to the Python-formatted processed output produced by PHITS Tools</li>\n\t\t<li><strong>plots</strong> refers to the <em>*.eps</em> files produced by PHITS and the <em>*.png</em> and <em>*.pdf</em> files produced by PHITS Tools, all containing graphical plots of tally output (only relevant to <strong>1E8</strong> simulations)</li>\n\t</ul>\n\t</li>\n\t<li>Other labels:\n\t<ul>\n\t\t<li><strong>proton-tally</strong>&nbsp;refers to output from the huge <em>[T-Cross]</em> tally used only in <strong>1E8</strong>&nbsp;simulations for scoring proton phase space as a function of energy, position, and direction (separated from others owing to its considerable size)</li>\n\t\t<li><strong>neutron-dump</strong> refers to the event-by-event neutron production data scored by a <em>[T-Product]</em> tally&#39;s &quot;dump&quot; option</li>\n\t\t<li><strong>NumPy</strong> and <strong>Pandas</strong>&nbsp;to denote if <strong>processed</strong>&nbsp;contents are formatted as NumPy record arrays or Pandas Dataframes&nbsp;</li>\n\t\t<li><strong>gamma-dump</strong> refers to the event-by-event gamma-ray production data scored by a <em>[T-Product]</em> tally&#39;s &quot;dump&quot; option</li>\n\t\t<li><strong>NumPy</strong>&nbsp;and <strong>Pandas</strong> to denote if <strong>processed</strong>&nbsp;contents are formatted as NumPy record arrays or Pandas Dataframes&nbsp;</li>\n\t\t<li><strong>other</strong> refers to output from all other tallies aside from the above three (energy deposition, LET, diagnostic tallies, etc.; only relevant to <strong>1E8</strong>&nbsp;simulations given all tallies except dump tallies were disabled for <strong>1E9</strong> simulations) along with (for <strong>raw</strong>&nbsp;directories) PHITS input-related files and <em>phits*.out</em> file(s).</li>\n\t</ul>\n\t</li>\n</ul>\n\n<p>All put together, this results in the following top-level directories contained in this repository:</p>\n\n<ul>\n\t<li><em>common_inputs</em></li>\n\t<li><em>GaussE_1E8_raw_proton-tally</em></li>\n\t<li><em>GaussE_1E8_raw_neutron-dump</em></li>\n\t<li><em>GaussE_1E8_raw_gamma-dump</em></li>\n\t<li><em>GaussE_1E8_raw_other</em></li>\n\t<li><em>GaussE_1E9_raw_neutron-dump&nbsp;<strong>(upon request)</strong></em></li>\n\t<li><em>GaussE_1E9_raw_gamma-dump&nbsp;<strong>(upon request)</strong></em></li>\n\t<li><em>GaussE_1E9_raw_other</em></li>\n\t<li><em>GaussE_1E8_processed_proton-tally</em></li>\n\t<li><em>GaussE_1E8_processed_neutron-dump_NumPy</em></li>\n\t<li><em>GaussE_1E8_processed_neutron-dump_Pandas</em></li>\n\t<li><em>GaussE_1E8_processed_gamma-dump_NumPy</em></li>\n\t<li><em>GaussE_1E8_processed_gamma-dump_Pandas</em></li>\n\t<li><em>GaussE_1E8_processed_other</em></li>\n\t<li><em>GaussE_1E9_processed_neutron-dump_NumPy <strong>(upon request)</strong></em></li>\n\t<li><em>GaussE_1E9_processed_neutron-dump_Pandas<strong>&nbsp;(upon request)</strong></em></li>\n\t<li><em>GaussE_1E9_processed_gamma-dump_NumPy<strong>&nbsp;(upon request)</strong></em></li>\n\t<li><em>GaussE_1E9_processed_gamma-dump_Pandas&nbsp;<strong>(upon request)</strong></em></li>\n\t<li><em>GaussE_1E9_processed_other&nbsp;</em></li>\n\t<li><em>GaussE_1E8_plots</em></li>\n\t<li><em>MonoE_1E8_raw_proton-tally</em></li>\n\t<li><em>MonoE_1E8_raw_neutron-dump</em></li>\n\t<li><em>MonoE_1E8_raw_gamma-dump</em></li>\n\t<li><em>MonoE_1E8_raw_other</em></li>\n\t<li><em>MonoE_1E9_raw_neutron-dump<strong>&nbsp;(upon request)</strong></em></li>\n\t<li><em>MonoE_1E9_raw_gamma-dump&nbsp;<strong>(upon request)</strong></em></li>\n\t<li><em>MonoE_1E9_raw_other</em></li>\n\t<li><em>MonoE_1E8_processed_proton-tally</em></li>\n\t<li><em>MonoE_1E8_processed_neutron-dump_NumPy</em></li>\n\t<li><em>MonoE_1E8_processed_neutron-dump_Pandas</em></li>\n\t<li><em>MonoE_1E8_processed_gamma-dump_NumPy</em></li>\n\t<li><em>MonoE_1E8_processed_gamma-dump_Pandas</em></li>\n\t<li><em>MonoE_1E8_processed_other</em></li>\n\t<li><em>MonoE_1E9_processed_neutron-dump_NumPy&nbsp;<strong>(upon request)</strong></em></li>\n\t<li><em>MonoE_1E9_processed_neutron-dump_Pandas&nbsp;<strong>(upon request)</strong></em></li>\n\t<li><em>MonoE_1E9_processed_gamma-dump_NumPy&nbsp;<strong>(upon request)</strong></em></li>\n\t<li><em>MonoE_1E9_processed_gamma-dump_Pandas<strong>&nbsp;(upon request)</strong></em></li>\n\t<li><em>MonoE_1E9_processed_other</em></li>\n\t<li><em>MonoE_1E8_plots</em></li>\n</ul>\n\n<p>And, as stated earlier, each of these top-level directories is divided into a <em>training</em> subdirectory (containing 37 <em>???_MeV</em> directories) and a <em>testing</em> subdirectory (containing 10 <em>???p?_MeV </em>directories), where the <em>???[p?]_MeV</em> directories only (1) contain particular files (2) relevant to certain simulations&amp;mdash;as specified by the top-level directory&#39;s name.</p>\n\n<p>&nbsp;</p>\n\n<p>As a note to anyone surveying the <em>raw</em> files, all <em>GaussE</em> simulations were ran with OpenMP parallelization with 10 processes.<br>\nFor<em> 1E8 </em>simulations, this was conducted as ten PHITS runs of 1E7 protons each; for<em> 1E9 </em>simulations, this was conducted as twenty runs of 5E7 protons each.<br>\n(PHITS runs can be &quot;chained&quot; as &quot;restart calculations&quot;, where one run can resume from where a previous run ended.)<br>\nIn these simulations, the generated<em> phits.out</em> files from each run were renamed to <em>phits-#.out</em> (where <em>#</em> is the run number,<em> 0</em> to<em> 19</em>) and moved into a <em>phitsout</em> subdirectory after each run&#39;s completion.<br>\nHowever, this was less uniform for the <em>MonoE</em> simulations; for those, the strategy was to complete each simulation in a single run of PHITS.&nbsp;<br>\nThis generally involved using a hybrid OpenMP + MPI parallelization with anywhere from 80 to 160 processes each, split between OMP and MPI (noting that some<em> 1E9 </em>runs were conducted with only MPI parallelization).<br>\nNone of this influences the output format of the standard tally outputs.<br>\nHowever, the number of dump files produced is equal to the number of MPI processes utilized.<br>\nThis means that each <em>GaussE</em> simulation only has one dump file per dump tally owing to only using OpenMP parallelization (which merges its dump files at the end of calculation) while the<em> MonoE </em>simulations contain a varied number of dump files per dump tally owing to varriations in parallelization strategies employed in those simulations.<br>\nPHITS Tools ultimately merges all dump outputs back together in its processing, meaning if looking at the <em>processed</em> output this quirk of how simulations were conducted should not be apparent at all.</p>\n\n<p><br>\nGiven PHITS Tools was under ongoing development as this dataset was being produced, the <em>GaussE</em> directories contain some extra output not present in the <em>MonoE</em> directories.&nbsp; Most notably, only for the <em>GaussE</em> simulations do the <em>plot</em> directories contain PNG and PDF plot files generated by PHITS Tools and the <em>*_processed_*</em> directories contain dictionary objects of the processed <em>phits*.out</em> files.</p>\n\n<p>Note that, for convenience, the <em>phits*.out</em> file(s) for each simulation are also copied to all <em>*_raw_*</em> directories.&nbsp; The <em>phits*.out</em> file(s) contain the full PHITS input echo, among other information about the simulation.&nbsp; For the <em>GaussE</em> simulations, these are within a further <em>phitsout</em> subdirectory for each beam energy.&nbsp; Also for all <em>GaussE_*_processed_*</em> directories, the processed <em>phits*.out</em> file(s), <em>phits*_out.pickle.xz</em>, are included too.</p>\n\n<p><strong>References</strong></p>\n\n<p><em>TO BE POPULATED</em></p>\n\n<p><strong>Acknowledgements</strong></p>\n\n<p>The NOVO project has received funding from the European Innovation Council (EIC) under grant agreement No. 101130979. The EIC receives support from the European Union&#39;s Horizon Europe research and innovation programme. Partners from The University of Manchester has received funding from UK Research and Innovation under grant agreement No. 10102118&nbsp;</p>", 
  "creator": [
    {
      "@id": "https://orcid.org/0000-0003-3761-5415", 
      "name": "Ratliff, Hunter", 
      "@type": "Person", 
      "affiliation": "Department of Computer science, Electrical engineering and Mathematical sciences, Western Norway University of Applied Sciences, Inndalsveien 28, Bergen, 5063, Vestland, Norway"
    }, 
    {
      "@id": "https://orcid.org/0009-0006-7725-6751", 
      "name": "Blangiardi, Francesco", 
      "@type": "Person", 
      "affiliation": "Technology Methods and Systems Data Based Methods, Fraunhofer ENAS, Technologie Campus 3, Chemnitz, 09126, Saxony, Germany"
    }, 
    {
      "@id": "https://orcid.org/0000-0002-9501-0898", 
      "name": "K\u00f6gler, Toni", 
      "@type": "Person", 
      "affiliation": "Helmholtz-Zentrum Dresden - Rossendorf"
    }
  ]
}