{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T08:05:21Z","timestamp":1789977921471,"version":"4.0.1"},"reference-count":14,"publisher":"IOP Publishing","issue":"5","license":[{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Biomed. Phys. Eng. Express"],"published-print":{"date-parts":[[2026,10,1]]},"abstract":"Abstract<\/jats:title>\n \n Objective.<\/jats:italic>\n To determine whether simple regularized linear regression models, relying solely on radial distance, mass density and elemental tissue composition as input features, can outperform the conventional water-kernel interpolation baseline for predicting radial dose point kernels (DPKs) of\n \n \n \n <\/jats:tex-math>\n \n \n \n \n \n 177<\/mml:mn>\n <\/mml:mrow>\n <\/mml:msup>\n \n Lu<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n across a diverse set of homogeneous tissues and one additional low-density homogeneous medium.\n Approach.<\/jats:italic>\n Radial DPKs for\n \n \n \n <\/jats:tex-math>\n \n \n \n \n \n 177<\/mml:mn>\n <\/mml:mrow>\n <\/mml:msup>\n \n Lu<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n were generated using OpenGATE 10 Monte Carlo simulations in 12 homogeneous media covering soft tissues, bone-like structures, and low-density lung-like tissues, together with one additional low-density homogeneous medium (\n \n \n \n <\/jats:tex-math>\n \n \n \u03c1<\/mml:mi>\n =<\/mml:mo>\n 0.296<\/mml:mn>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n g\u2009cm\n \n \n \n <\/jats:tex-math>\n \n \n \n \n \n \u2212<\/mml:mo>\n 3<\/mml:mn>\n <\/mml:mrow>\n <\/mml:msup>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n ) evaluated separately as an out-of-distribution extrapolation test. Three scikit-learn models (RidgeCV, LassoCV, and ElasticNetCV) were trained with robust scaling and evaluated through leave-one-tissue-out (LOTO) and leave-one-family-out cross-validation, as well as on the independent low-density case. Performance was assessed using mean absolute percentage error on physical dose (\n \n \n \n <\/jats:tex-math>\n \n \n \n \n MAPE<\/mml:mi>\n <\/mml:mrow>\n \n \n phys<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:msub>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n ), logarithmic root mean square error (\n \n \n \n <\/jats:tex-math>\n \n \n \n \n RMSE<\/mml:mi>\n <\/mml:mrow>\n \n log<\/mml:mi>\n <\/mml:mrow>\n <\/mml:msub>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n ), relative integral energy error, and relative errors on\n \n \n \n <\/jats:tex-math>\n \n \n \n r<\/mml:mi>\n \n 50<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:msub>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n and\n \n \n \n <\/jats:tex-math>\n \n \n \n r<\/mml:mi>\n \n 10<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:msub>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n , systematically compared against a water-kernel baseline obtained by linear resampling of a single simulated water DPK.\n Main results.<\/jats:italic>\n ElasticNetCV was the best-performing linear surrogate overall, achieving a mean\n \n \n \n <\/jats:tex-math>\n \n \n \n \n MAPE<\/mml:mi>\n <\/mml:mrow>\n \n \n phys<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:msub>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n of 224% in LOTO validation (median 117%), followed by LassoCV and RidgeCV. However, none of the linear surrogates outperformed the water-kernel baseline on the main physical performance metrics. Mean gains relative to the baseline remained negative across tissue families, ranging from about\n \n \n \n <\/jats:tex-math>\n \n \n \u2212<\/mml:mo>\n 69<\/mml:mn>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n percentage points in bone-like tissues to about\n \n \n \n <\/jats:tex-math>\n \n \n \u2212<\/mml:mo>\n 759<\/mml:mn>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n percentage points in low-density lung-like tissues. Relative integral energy errors reached several thousand percent in lung-like media, and the separate low-density homogeneous medium also showed severe degradation (best surrogate\n \n \n \n <\/jats:tex-math>\n \n \n \n \n MAPE<\/mml:mi>\n <\/mml:mrow>\n \n \n phys<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:msub>\n \u2248<\/mml:mo>\n 494<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n versus\n \n \n \n <\/jats:tex-math>\n \n \n \u2248<\/mml:mo>\n 45<\/mml:mn>\n %<\/mml:mi>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n for the water-kernel baseline). Although some surrogate predictions showed lower log-domain errors than the baseline in selected cases, these improvements did not translate into better dosimetric accuracy in physical dose units.\n Significance.<\/jats:italic>\n Simple linear machine-learning surrogates based on radial distance, mass density and macroscopic elemental composition did not improve, and often substantially worsened, the accuracy of radial\n \n \n \n <\/jats:tex-math>\n \n \n \n \n \n 177<\/mml:mn>\n <\/mml:mrow>\n <\/mml:msup>\n \n Lu<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:math>\n <\/jats:inline-formula>\n DPK prediction compared with standard water-kernel interpolation. This negative benchmark delineates the limitations of basic tabular linear regression for capturing density- and composition-dependent kernel changes across a range of homogeneous media relevant to nuclear medicine dosimetry, and provides a reproducible reference framework for future development of more advanced surrogate models.\n <\/jats:p>","DOI":"10.1088\/2057-1976\/aea427","type":"journal-article","created":{"date-parts":[[2026,9,8]],"date-time":"2026-09-08T22:48:17Z","timestamp":1788907697000},"page":"055019","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Limitations of simple machine-learning surrogates for cross-tissue prediction of radial\n \n \n \n \n \n 177<\/mml:mn>\n <\/mml:mrow>\n <\/mml:msup>\n \n Lu<\/mml:mi>\n <\/mml:mrow>\n <\/mml:mrow>\n <\/mml:math>\n dose point kernels"],"prefix":"10.1088","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-2734-4948","authenticated-orcid":true,"given":"Houcine","family":"Ennassiri","sequence":"first","affiliation":[{"name":"Centre Hospitalier Intercommunal Toulon\u2013La Seyne-sur-Mer"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,9,21]]},"reference":[{"key":"bpexaea427bib1","doi-asserted-by":"publisher","first-page":"125","DOI":"10.1016\/S0969-8051(98)00087-1","type":"journal-article","article-title":"Radiation dosimetry in nuclear medicine","volume":"26","author":"Stabin","year":"1999","journal-title":"Nucl. Med. Biol."},{"key":"bpexaea427bib2","doi-asserted-by":"publisher","first-page":"151","DOI":"10.2967\/jnumed.115.159012","type":"journal-article","article-title":"MIRD pamphlet No. 26: joint EANM\/MIRD guidelines for quantitative 177Lu SPECT applied for dosimetry of radiopharmaceutical therapy","volume":"57","author":"Ljungberg","year":"2016","journal-title":"J. Nucl. Med."},{"key":"bpexaea427bib3","doi-asserted-by":"publisher","first-page":"11S","DOI":"10.2967\/jnumed.40.1.11S","type":"journal-article","article-title":"MIRD pamphlet No. 17: the dosimetry of nonuniform activity distributions\u2014radionuclide S values at the voxel level","volume":"40","author":"Bolch","year":"1999","journal-title":"J. Nucl. Med."},{"key":"bpexaea427bib4","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1088\/0031-9155\/40\/3\/003","type":"journal-article","article-title":"Validation of a dose-point kernel convolution technique for internal dosimetry","volume":"40","author":"Giap","year":"1995","journal-title":"Phys. Med. Biol."},{"key":"bpexaea427bib5","doi-asserted-by":"publisher","first-page":"5238","DOI":"10.1118\/1.4737096","type":"journal-article","article-title":"A dose point kernel database using GATE Monte Carlo simulation toolkit for nuclear medicine applications: comparison with other Monte Carlo codes","volume":"39","author":"Papadimitroulas","year":"2012","journal-title":"Med. Phys."},{"key":"bpexaea427bib6","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1089\/cbr.2016.2117","type":"journal-article","article-title":"Evaluating the application of tissue-specific dose kernels instead of water dose kernels in internal dosimetry: a Monte Carlo study","volume":"31","author":"Khazaee Moghadam","year":"2016","journal-title":"Cancer Biother. Radiopharm."},{"key":"bpexaea427bib7","doi-asserted-by":"publisher","first-page":"236","DOI":"10.2967\/jnumed.112.105825","type":"journal-article","article-title":"Study of the impact of tissue density heterogeneities on 3-dimensional abdominal dosimetry: comparison between dose kernel convolution and direct Monte Carlo methods","volume":"54","author":"Dieudonn\u00e9","year":"2013","journal-title":"J. Nucl. Med."},{"key":"bpexaea427bib8","doi-asserted-by":"publisher","first-page":"531","DOI":"10.1667\/RR15563.1","type":"journal-article","article-title":"The impact of tissue type and density on dose point kernels for patient-specific voxel-wise dosimetry: a Monte Carlo investigation","volume":"193","author":"Tiwari","year":"2020","journal-title":"Radiat. Res."},{"key":"bpexaea427bib9","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac8c83","type":"journal-article","article-title":"The OpenGATE ecosystem for Monte Carlo simulation in medical physics","volume":"67","author":"Sarrut","year":"2022","journal-title":"Phys. Med. Biol."},{"key":"bpexaea427bib10","doi-asserted-by":"publisher","first-page":"4543","DOI":"10.1088\/0031-9155\/49\/19\/007","type":"journal-article","article-title":"GATE: a simulation toolkit for PET and SPECT","volume":"49","author":"Jan","year":"2004","journal-title":"Phys. Med. Biol."},{"key":"bpexaea427bib11","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1186\/s40658-023-00560-9","type":"journal-article","article-title":"A machine learning-based model for a dose point kernel calculation","volume":"10","author":"Scarinci","year":"2023","journal-title":"EJNMMI Phys."},{"key":"bpexaea427bib12","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.21790170","type":"other","article-title":"Radial 177Lu dose point kernels in 13 media: OpenGATE simulations, profiles and surrogate-model analysis [Data set]","author":"Ennassiri","year":"2026"},{"key":"bpexaea427bib13","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1016\/j.mibio.2004.03.002","type":"journal-article","article-title":"Development of a 4-D digital mouse phantom for molecular imaging research","volume":"6","author":"Segars","year":"2004","journal-title":"Mol. Imaging Biol."},{"key":"bpexaea427bib14","doi-asserted-by":"publisher","first-page":"2825","DOI":"10.5555\/1953048.2078195","type":"journal-article","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."}],"container-title":["Biomedical Physics & Engineering Express"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T07:25:38Z","timestamp":1789975538000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/aea427"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,21]]},"references-count":14,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,9,21]]},"published-print":{"date-parts":[[2026,10,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2057-1976\/aea427","relation":{},"ISSN":["2057-1976"],"issn-type":[{"value":"2057-1976","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,21]]},"assertion":[{"value":"Limitations of simple machine-learning surrogates for cross-tissue prediction of radial\n \n \n \n \n \n 177\n \n \n \n Lu\n \n \n \n dose point kernels","name":"article_title","label":"Article Title"},{"value":"Biomedical Physics & Engineering Express","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2026-04-07","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-09-08","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-09-21","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}