{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T12:31:32Z","timestamp":1777984292494,"version":"3.51.4"},"reference-count":29,"publisher":"SAGE Publications","license":[{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"DOI":"10.13039\/100000098","name":"Clinical Center","doi-asserted-by":"publisher","award":["P30 CA008748"],"award-info":[{"award-number":["P30 CA008748"]}],"id":[{"id":"10.13039\/100000098","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Technol Cancer Res Treat"],"published-print":{"date-parts":[[2026,4]]},"abstract":"\n Introduction<\/jats:title>\n Although long short-term memory (LSTM) networks have been tested to predict short-term respiratory motion, their performance in long-term forecasting under breathing irregularities must be assessed. We aim to evaluate and enhance the long-term prediction of internal motion from an external surrogate using subject-specific LSTM models through a hybrid, adaptive approach.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n \n Concurrent internal navigator and external bellows respiratory-motion waveforms were acquired for ten volunteers during two four-dimensional magnetic resonance imaging (4DMRI) scans lasting 3\u201310\u2005min each. Approximately 20\u2005min intervened between the first (mid-term) and second (long-term) scan. After training on the first half of the mid-term data, subject-specific LSTM models were applied to the remaining mid-term and entire long-term datasets to predict internal waveforms. The accuracy of a model's prediction was assessed with Pearson's correlation (\n C<\/jats:italic>\n ), referenced to the native waveforms, maximized through the time-domain cross-correlation (TCC), and enhanced by correcting residual phase shifts in the LSTM models using a hybrid (LSTM-TCC) approach. Hyperparameter selection by minimizing the root mean square error (RMSE) to identify high-performance (\n C<\/jats:italic>\n \u2009\u2265\u20090.8) LSTM models was evaluated by the area under the receiver operating characteristic curve (AUROC). The temporal accuracy of inspiratory-peak predictions was characterized.\n <\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Compared to the native waveforms (\n C<\/jats:italic>\n \u2009=\u20090.42\u2009\u00b1\u20090.28) and TCC method (\n C<\/jats:italic>\n \u2009=\u20090.77\u2009\u00b1\u20090.09), the LSTM models yielded more accurate predictions (\n C<\/jats:italic>\n \u2009=\u20090.89\u2009\u00b1\u20090.07) in the mid-term scans. Over 20\u201330\u2005min, LSTM predictions faltered (\n C<\/jats:italic>\n \u2009<\u20090.80) in two subjects but were rescued by LSTM-TCC (\n C<\/jats:italic>\n \u2009=\u20090.90\u2009\u00b1\u20090.09). The temporal error in predicting inspiratory peaks was smaller for LSTM-TCC (\u0394\n t<\/jats:italic>\n \u2009=\u20090.15\u2009\u00b1\u20090.11sec) than LSTM (\u0394\n t<\/jats:italic>\n \u2009=\u20090.18\u2009\u00b1\u20090.15sec). RMSE reliably identified high-performance models:\n \n \n A<\/mml:mi>\n U<\/mml:mi>\n R<\/mml:mi>\n O<\/mml:mi>\n \n C<\/mml:mi>\n \n L<\/mml:mi>\n S<\/mml:mi>\n T<\/mml:mi>\n M<\/mml:mi>\n <\/mml:mrow>\n \n m<\/mml:mi>\n i<\/mml:mi>\n d<\/mml:mi>\n \u2212<\/mml:mo>\n t<\/mml:mi>\n e<\/mml:mi>\n r<\/mml:mi>\n m<\/mml:mi>\n <\/mml:mrow>\n <\/mml:msubsup>\n =<\/mml:mo>\n 0.82<\/mml:mn>\n <\/mml:math>\n <\/jats:inline-formula>\n ,\n \n \n A<\/mml:mi>\n U<\/mml:mi>\n R<\/mml:mi>\n O<\/mml:mi>\n \n C<\/mml:mi>\n \n L<\/mml:mi>\n S<\/mml:mi>\n T<\/mml:mi>\n M<\/mml:mi>\n <\/mml:mrow>\n \n l<\/mml:mi>\n o<\/mml:mi>\n n<\/mml:mi>\n g<\/mml:mi>\n \u2212<\/mml:mo>\n t<\/mml:mi>\n e<\/mml:mi>\n r<\/mml:mi>\n m<\/mml:mi>\n <\/mml:mrow>\n <\/mml:msubsup>\n =<\/mml:mo>\n 0.74<\/mml:mn>\n <\/mml:math>\n <\/jats:inline-formula>\n , and\n \n \n A<\/mml:mi>\n U<\/mml:mi>\n R<\/mml:mi>\n O<\/mml:mi>\n \n C<\/mml:mi>\n \n h<\/mml:mi>\n y<\/mml:mi>\n b<\/mml:mi>\n r<\/mml:mi>\n i<\/mml:mi>\n d<\/mml:mi>\n <\/mml:mrow>\n \n l<\/mml:mi>\n o<\/mml:mi>\n n<\/mml:mi>\n g<\/mml:mi>\n \u2212<\/mml:mo>\n t<\/mml:mi>\n e<\/mml:mi>\n r<\/mml:mi>\n m<\/mml:mi>\n <\/mml:mrow>\n <\/mml:msubsup>\n =<\/mml:mo>\n 0.83<\/mml:mn>\n <\/mml:math>\n <\/jats:inline-formula>\n .\n <\/jats:p>\n <\/jats:sec>\n \n Conclusion<\/jats:title>\n The feasibility of a novel adaptive subject-specific LSTM-TCC modeling was tested in 10 subjects, demonstrating that high accuracy of external-to-internal motion predictions in 3\u201310\u2005min can be extended to 30\u2005min overcoming breathing irregularities without remodeling. Further investigations of the adaptive LSTM-TCC model are warranted as a potential clinical solution.<\/jats:p>\n <\/jats:sec>","DOI":"10.1177\/15330338261420917","type":"journal-article","created":{"date-parts":[[2026,4,15]],"date-time":"2026-04-15T15:04:35Z","timestamp":1776265475000},"update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["A Feasibility Study of Hybrid Deep-Learning Prediction with Online Adaptation of Breathing Irregularities for Long-Term Internal Organ Motion During Radiotherapy"],"prefix":"10.1177","volume":"25","author":[{"given":"Andrew","family":"Milewski","sequence":"first","affiliation":[{"name":"Memorial Sloan Kettering Cancer Center"},{"name":"Department of Anesthesiology, Weill Cornell Medicine, New York, NY, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Uddin","family":"Fayed","sequence":"additional","affiliation":[{"name":"Memorial Sloan Kettering Cancer Center"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Vyas","family":"Gupta","sequence":"additional","affiliation":[{"name":"Memorial Sloan Kettering Cancer Center"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xingyu","family":"Nie","sequence":"additional","affiliation":[{"name":"Kentucky University College of Medicine"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9022-2883","authenticated-orcid":false,"given":"Guang","family":"Li","sequence":"additional","affiliation":[{"name":"Memorial Sloan Kettering Cancer Center"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","published-online":{"date-parts":[[2026,4,15]]},"reference":[{"key":"e_1_3_5_2_2","doi-asserted-by":"publisher","DOI":"10.1118\/1.2349696"},{"key":"e_1_3_5_3_2","first-page":"374","volume-title":"Treatment planning in radiation therapy","author":"Li G","year":"2021","unstructured":"Li G, Mageras GS, Dong L, Mohan R. Image-guided radiation therapy. In: Khan FM, Sperduto PW, Gibbons JP, eds. Treatment planning in radiation therapy. Wolters Kluwer; 2021:374-403."},{"key":"e_1_3_5_4_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2008.08.036"},{"key":"e_1_3_5_5_2","doi-asserted-by":"publisher","DOI":"10.1088\/0031-9155\/52\/16\/004"},{"key":"e_1_3_5_6_2","doi-asserted-by":"publisher","DOI":"10.1118\/1.3455276"},{"key":"e_1_3_5_7_2","doi-asserted-by":"publisher","DOI":"10.1118\/1.3582693"},{"key":"e_1_3_5_8_2","doi-asserted-by":"publisher","DOI":"10.1118\/1.4904023"},{"key":"e_1_3_5_9_2","doi-asserted-by":"publisher","DOI":"10.1002\/mp.12447"},{"key":"e_1_3_5_10_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2019.02.001"},{"key":"e_1_3_5_11_2","doi-asserted-by":"publisher","DOI":"10.1177\/15330338221111592"},{"key":"e_1_3_5_12_2","doi-asserted-by":"publisher","DOI":"10.1177\/153303461000900603"},{"key":"e_1_3_5_13_2","doi-asserted-by":"publisher","DOI":"10.21037\/qims-19-1058"},{"key":"e_1_3_5_14_2","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14140"},{"key":"e_1_3_5_15_2","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2021.3082578"},{"key":"e_1_3_5_16_2","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13748"},{"key":"e_1_3_5_17_2","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15321"},{"key":"e_1_3_5_18_2","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ab13fa"},{"key":"e_1_3_5_19_2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0275719"},{"key":"e_1_3_5_20_2","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-020-01729-7"},{"key":"e_1_3_5_21_2","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13717"},{"key":"e_1_3_5_22_2","doi-asserted-by":"publisher","DOI":"10.2196\/27235"},{"key":"e_1_3_5_23_2","doi-asserted-by":"publisher","DOI":"10.1186\/s13014-024-02532-4"},{"key":"e_1_3_5_24_2","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/ac60b7"},{"key":"e_1_3_5_25_2","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2022.898771"},{"key":"e_1_3_5_26_2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3117837"},{"key":"e_1_3_5_27_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2016.11.004"},{"key":"e_1_3_5_28_2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.g7594"},{"key":"e_1_3_5_29_2","doi-asserted-by":"publisher","DOI":"10.1118\/1.1558675"},{"key":"e_1_3_5_30_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2012.05.006"}],"container-title":["Technology in Cancer Research & Treatment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/15330338261420917","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/15330338261420917","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/15330338261420917","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T06:17:37Z","timestamp":1777702657000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/15330338261420917"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":29,"alternative-id":["10.1177\/15330338261420917"],"URL":"https:\/\/doi.org\/10.1177\/15330338261420917","relation":{"has-review":[{"id-type":"doi","id":"10.1177\/15330338261420917\/v1\/review2","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v1\/review1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v2\/response1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v1\/review3","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v1\/decision1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v3\/decision1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v2\/review2","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v2\/review1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v4\/review1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v4\/decision1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v4\/response1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v3\/review1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v3\/response1","asserted-by":"object"},{"id-type":"doi","id":"10.1177\/15330338261420917\/v2\/decision1","asserted-by":"object"}]},"ISSN":["1533-0346","1533-0338"],"issn-type":[{"value":"1533-0346","type":"print"},{"value":"1533-0338","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4]]},"article-number":"15330338261420917"}}