{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T10:26:20Z","timestamp":1787567180490,"version":"build-2736575974"},"reference-count":36,"publisher":"IOP Publishing","issue":"16","license":[{"start":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T00:00:00Z","timestamp":1787529600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T00:00:00Z","timestamp":1787529600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/100000070","name":"National Institute of Biomedical Imaging and Bioengineering","doi-asserted-by":"crossref","award":["R01 EB034475"],"award-info":[{"award-number":["R01 EB034475"]}],"id":[{"id":"10.13039\/100000070","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Phys. Med. Biol."],"published-print":{"date-parts":[[2026,8,28]]},"abstract":"Abstract<\/jats:title>\n \n Objective.<\/jats:italic>\n Monte Carlo simulations of optical photon transport are computationally prohibitive for large-scale optical systems including detector arrays and positron emission tomography systems, restricting their practical use to single-crystal studies. This work presents an enhanced conditional generative adversarial network capable of replacing optical simulations at the crystal array level, extending our previous single-crystal approach to a 3\u2009\n \n \n \n <\/jats:tex-math>\n \n \n \u00d7<\/mml:mo>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n \u20093 bismuth germanate detector array.\n Approach.<\/jats:italic>\n We introduce Fourier feature encoding and a learnable latent mapping network as the modifications enabling stable training on the array geometry, together with a physics-informed loss term enforcing the unit-sphere state space\n \n \n \n <\/jats:tex-math>\n \n \n \n S<\/mml:mi>\n 2<\/mml:mn>\n <\/mml:msup>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n of the generated propagation directions as a soft constraint. Training data requirements are reduced eight-fold by exploiting the array\u2019s symmetry. The model is evaluated in three studies: a full array evaluation, a high-resolution study probing generalization, and a pencil beam irradiation study assessing practical applicability. Performance is benchmarked against GATE10\/Geant4 ground truth, using the fluctuations between independent Monte Carlo runs.\n Main results.<\/jats:italic>\n The enhanced optiGAN achieves similarity values within\n \n \n \n <\/jats:tex-math>\n \n \n 3<\/mml:mn>\n \u03c3<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n -agreement of the Monte Carlo baseline across all evaluation conditions. An ablation and attribution analysis shows that the physics-informed loss term reduces low-structural similarity index measure bin fractions by a factor of 3.6 on the outer crystals, with a localized trade-off at the central crystal, yielding a net 48% reduction over the full array. The model transitions from electron-emission training data to realistic\n \n \n \n <\/jats:tex-math>\n \n \n \u03b3<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n -photon interactions, producing flood maps that reproduce experimental patterns including photopeak clusters and inter-crystal scatter lines.\n Significance.<\/jats:italic>\n This proof-of-concept demonstrates that a physics-informed generative model can simulate optical photon transport in segmented scintillator arrays at a training and inference cost accessible on a single workstation GPU. This provides a foundation for future models capable of generalizing across diverse array configurations.\n <\/jats:p>","DOI":"10.1088\/1361-6560\/ae95d7","type":"journal-article","created":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T22:57:58Z","timestamp":1785970678000},"page":"165022","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["OptiGAN for crystal arrays: physics-informed generative modeling of optical photon transport in PET detector arrays"],"prefix":"10.1088","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0306-7641","authenticated-orcid":true,"given":"Stephan","family":"Naunheim","sequence":"first","affiliation":[{"name":"University of California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6483-9858","authenticated-orcid":false,"given":"Brandon","family":"Pardi","sequence":"additional","affiliation":[{"name":"University of California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-5745-7249","authenticated-orcid":false,"given":"Guneet","family":"Mummaneni","sequence":"additional","affiliation":[{"name":"University of California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0414-9369","authenticated-orcid":true,"given":"Carlotta","family":"Trigila","sequence":"additional","affiliation":[{"name":"University of California"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2439-1064","authenticated-orcid":true,"given":"Emilie","family":"Roncali","sequence":"additional","affiliation":[{"name":"University of California"},{"name":"University of California"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,8,24]]},"reference":[{"key":"pmbae95d7bib1","article-title":"Wasserstein GAN","author":"Arjovsky","year":"2017","type":"other"},{"key":"pmbae95d7bib2","author":"Bailey","year":"2005","type":"book"},{"key":"pmbae95d7bib3","doi-asserted-by":"publisher","first-page":"311","DOI":"10.1053\/j.semnuclmed.2018.02.006","type":"journal-article","article-title":"Innovations in instrumentation for positron emission tomography","volume":"48","author":"Berg","year":"2018","journal-title":"Semin. Nucl. Med."},{"key":"pmbae95d7bib4","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1109\/JSEN.2015.2483565","type":"journal-article","article-title":"SPAD figures of merit for photon-counting, photon-timing and imaging applications: a review","volume":"16","author":"Bronzi","year":"2016","journal-title":"IEEE Sens. J."},{"key":"pmbae95d7bib5","author":"Cherry","year":"2012","edition":"4th edn.","type":"book"},{"key":"pmbae95d7bib6","doi-asserted-by":"publisher","first-page":"856","DOI":"10.1002\/mp.17484","type":"journal-article","article-title":"Algorithms to reduce the edge effect and improve the flood histogram quality of a PET detector consisting of two pixelated crystal arrays","volume":"52","author":"Cong","year":"2025","journal-title":"Med. Phys."},{"key":"pmbae95d7bib7","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1007\/s41781-017-0004-6","type":"journal-article","article-title":"Learning particle physics by example: location-aware generative adversarial networks for physics synthesis","volume":"1","author":"de Oliveira","year":"2017","journal-title":"Comput. Softw. Big Sci."},{"key":"pmbae95d7bib8","doi-asserted-by":"publisher","first-page":"C02012","DOI":"10.1088\/1748-0221\/15\/02\/C02012","type":"journal-article","article-title":"Machine learning for imaging Cherenkov detectors","volume":"15","author":"Fanelli","year":"2020","journal-title":"J. Instrum."},{"key":"pmbae95d7bib9","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/ab845a","type":"journal-article","article-title":"DeepRICH: learning deeply Cherenkov detectors","volume":"1","author":"Fanelli","year":"2020","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"pmbae95d7bib10","article-title":"Generative adversarial networks","author":"Goodfellow","year":"2014","type":"other"},{"key":"pmbae95d7bib11","article-title":"Improved training of wasserstein GANs","author":"Gulrajani","year":"2017","type":"other"},{"key":"pmbae95d7bib12","first-page":"pp 1","type":"conference-proceedings","article-title":"Physics-informed neural networks for quantum eigenvalue problems","author":"Jin","year":"2022"},{"key":"pmbae95d7bib13","doi-asserted-by":"publisher","first-page":"422","DOI":"10.1038\/s42254-021-00314-5","type":"journal-article","article-title":"Physics-informed machine learning","volume":"3","author":"Karniadakis","year":"2021","journal-title":"Nat. Rev. Phys."},{"key":"pmbae95d7bib14","first-page":"pp 4396","type":"conference-proceedings","article-title":"A style-based generator architecture for generative adversarial networks","author":"Karras","year":"2019"},{"key":"pmbae95d7bib15","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ae237c","type":"journal-article","article-title":"GATE 10 Monte Carlo particle transport simulation -Part II: architecture and innovations","volume":"71","author":"Krah","year":"2025","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib16","article-title":"Conditional generative adversarial nets","author":"Mirza","year":"2014","type":"other"},{"key":"pmbae95d7bib17","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ade2b5","type":"journal-article","article-title":"optiGAN: a deep learning-based alternative to optical photon tracking in Python-based GATE (10+)","volume":"70","author":"Mummaneni","year":"2025","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib18","type":"conference-proceedings","article-title":"A novel OptiGAN model: conditional generation of optical photons in PET crystal arrays with momentum conservation","volume":"vol 67","author":"Naunheim","year":"2026"},{"key":"pmbae95d7bib19","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.120.042003","type":"journal-article","article-title":"Accelerating science with generative adversarial networks: an application to 3D particle showers in multilayer calorimeters","volume":"120","author":"Paganini","year":"2018","journal-title":"Phys. Rev. Lett."},{"key":"pmbae95d7bib20","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevD.97.014021","type":"journal-article","article-title":"CaloGAN: simulating 3D high energy particle showers in multilayer electromagnetic calorimeters with generative adversarial networks","volume":"97","author":"Paganini","year":"2018","journal-title":"Phys. Rev. D"},{"key":"pmbae95d7bib21","first-page":"210","type":"journal-article","article-title":"Application of annihilation coincidence detection to transaxial reconstruction tomography","volume":"16","author":"Phelps","year":"1975","journal-title":"J. Nucl. Med."},{"key":"pmbae95d7bib22","type":"conference-proceedings","article-title":"Random features for large-scale kernel machines","volume":"vol 20","author":"Rahimi","year":"2007"},{"key":"pmbae95d7bib23","first-page":"pp 1","type":"conference-proceedings","article-title":"Optimizing PET detector performance and calibration through flood histogram analysis","author":"Refaey","year":"2024"},{"key":"pmbae95d7bib24","doi-asserted-by":"publisher","first-page":"2185","DOI":"10.1088\/0031-9155\/58\/7\/2185","type":"journal-article","article-title":"Simulation of light transport in scintillators based on 3D characterization of crystal surfaces","volume":"58","author":"Roncali","year":"2013","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib25","doi-asserted-by":"publisher","DOI":"10.1088\/2057-1976\/ab0f93","type":"journal-article","article-title":"Cerenkov light transport in scintillation crystals explained: realistic simulation with GATE","volume":"5","author":"Roncali","year":"2019","journal-title":"Biomed. Phys. Eng. Express"},{"key":"pmbae95d7bib26","doi-asserted-by":"publisher","first-page":"10TR03","DOI":"10.1088\/1361-6560\/abf276","type":"journal-article","article-title":"Advanced Monte Carlo simulations of emission tomography imaging systems with GATE","volume":"66","author":"Sarrut","year":"2021","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib27","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ae237b","type":"journal-article","article-title":"GATE 10 Monte Carlo particle transport simulation -Part I: development and new features","volume":"71","author":"Sarrut","year":"2025","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib28","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/abde9a","type":"journal-article","article-title":"Modeling complex particles phase space with GAN for Monte Carlo SPECT simulations: a proof of concept","volume":"66","author":"Sarrut","year":"2021","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib29","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab3fc1","type":"journal-article","article-title":"Generative adversarial networks (GAN) for compact beam source modelling in Monte Carlo simulations","volume":"64","author":"Sarrut","year":"2019","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib30","first-page":"pp 7537","type":"conference-proceedings","article-title":"Fourier features let networks learn high frequency functions in low dimensional domains","volume":"vol 33","author":"Tancik","year":"2020"},{"key":"pmbae95d7bib31","doi-asserted-by":"publisher","first-page":"734","DOI":"10.1109\/TRPMS.2024.3395131","type":"journal-article","article-title":"Intercrystal optical crosstalk in radiation detectors: Monte Carlo modeling and experimental validation","volume":"8","author":"Trigila","year":"2024","journal-title":"IEEE Trans. Radiat. Plasma Med. Sci."},{"key":"pmbae95d7bib32","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/adde0c","type":"journal-article","article-title":"Towards large nuclear imaging system optical simulations with optiGAN, a generative adversarial network","volume":"70","author":"Trigila","year":"2025","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib33","doi-asserted-by":"publisher","first-page":"21NT03","DOI":"10.1088\/1361-6560\/ac2e18","type":"journal-article","article-title":"Integration of polarization in the LUTDavis model for optical Monte Carlo simulation in radiation detectors","volume":"66","author":"Trigila","year":"2021","journal-title":"Phys. Med. Biol."},{"key":"pmbae95d7bib34","doi-asserted-by":"publisher","DOI":"10.1088\/2632-2153\/acc782","type":"journal-article","article-title":"A generative adversarial network to speed up optical Monte Carlo simulations","volume":"4","author":"Trigila","year":"2023","journal-title":"Mach. Learn.: Sci. Technol."},{"key":"pmbae95d7bib35","doi-asserted-by":"publisher","first-page":"600","DOI":"10.1109\/TIP.2003.819861","type":"journal-article","article-title":"Image quality assessment: from error visibility to structural similarity","volume":"13","author":"Wang","year":"2004","journal-title":"IEEE Trans. Image Process."},{"key":"pmbae95d7bib36","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac0b82","type":"journal-article","article-title":"Design study of a PET detector with 0.5 mm crystal pitch for high-resolution preclinical imaging","volume":"66","author":"Zhang","year":"2021","journal-title":"Phys. Med. Biol."}],"container-title":["Physics in Medicine & Biology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T08:51:26Z","timestamp":1787561486000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1361-6560\/ae95d7"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,24]]},"references-count":36,"journal-issue":{"issue":"16","published-online":{"date-parts":[[2026,8,24]]},"published-print":{"date-parts":[[2026,8,28]]}},"URL":"https:\/\/doi.org\/10.1088\/1361-6560\/ae95d7","relation":{},"ISSN":["0031-9155","1361-6560"],"issn-type":[{"value":"0031-9155","type":"print"},{"value":"1361-6560","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,24]]},"assertion":[{"value":"OptiGAN for crystal arrays: physics-informed generative modeling of optical photon transport in PET detector arrays","name":"article_title","label":"Article Title"},{"value":"Physics in Medicine & Biology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 The Author(s). Published on behalf of Institute of Physics and Engineering in Medicine by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2026-02-02","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-05","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-24","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}