{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,2]],"date-time":"2026-10-02T02:07:01Z","timestamp":1790906821726,"version":"4.1.0"},"reference-count":22,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2025,4,22]],"date-time":"2025-04-22T00:00:00Z","timestamp":1745280000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/page\/copyright"},{"start":{"date-parts":[[2025,4,22]],"date-time":"2025-04-22T00:00:00Z","timestamp":1745280000000},"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":[[2025,5,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>\n                  <jats:italic>Purpose<\/jats:italic>. Traditional machine learning (ML) and deep learning (DL) applications in treatment planning rely on complex model architectures and large, high-quality training datasets. However, they cannot fully replace the conventional optimization process. This study presents a novel application of ML in treatment planning where established ML\/DL toolkits are directly applied to treatment plan optimization. <jats:italic>Materials and Methods<\/jats:italic>. A one-layer network was designed based on the dose deposition matrix and implemented in PyTorch\u2019s L-BFGS optimizer with GPU acceleration. The classical steepest descent optimizer was selected as a reference for comparison. Both optimizers utilized identical inputs and objective functions to ensure a fair comparison. DVH- and gEUD-based objectives were implemented in standard quadratic forms. Standard uniform and 1,000 random initializations were used to test optimizer\u2019s search ability under different starting conditions for prostate and head-and-neck cases. <jats:italic>Results<\/jats:italic>. The MLT-assisted framework demonstrated comparable or superior plan quality to classical optimization by achieving lower objective values, improved DVHs and capturing finer modulation details in fluence maps. For gEUD-based optimization, it effectively explored beam weight elevations that classical optimization could only reach with stricter convergence criteria and many more iterations. The quality differences primarily stemmed from convergence speed. The MLT-assisted framework required significantly fewer evaluations and iterations to achieve similar or better results. Optimization on random initial maps further demonstrated that it was more robust and less likely to be trapped. It does not require stricter convergence criteria or extended runs to reach high-quality optima, making it more efficient and reliable. <jats:italic>Conclusion<\/jats:italic>. This framework leverages ML toolkits in a novel way, enabling faster convergence, greater robustness and handling of complex constraints. As the first study of its kind, it establishes MLT-assisted optimization as a viable and effective alternative to classical methods.<\/jats:p>","DOI":"10.1088\/2057-1976\/adcaca","type":"journal-article","created":{"date-parts":[[2025,4,9]],"date-time":"2025-04-09T22:47:39Z","timestamp":1744238859000},"page":"035016","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["A machine learning toolkit assisted approach for IMRT fluence map optimization: feasibility and advantages"],"prefix":"10.1088","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1108-7626","authenticated-orcid":true,"given":"Xin","family":"Wu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4559-5701","authenticated-orcid":true,"given":"Dongrong","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yang","family":"Sheng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qing-Rong Jackie","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0748-7280","authenticated-orcid":true,"given":"Qiuwen","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2025,4,22]]},"reference":[{"key":"bpexadcacabib1","first-page":"265","article-title":"TensorFlow: a system for large-scale machine learning","author":"Abadi","year":"2016"},{"key":"bpexadcacabib2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13014-024-02401-0","article-title":"Automatic IMRT treatment planning through fluence prediction and plan fine-tuning for nasopharyngeal carcinoma","volume":"19","author":"Cai","year":"2024","journal-title":"Radiation Oncology"},{"key":"bpexadcacabib3","doi-asserted-by":"publisher","first-page":"6148","DOI":"10.1002\/mp.12621","article-title":"Neural network dose models for knowledge-based planning in pancreatic SBRT","volume":"44","author":"Campbell","year":"2017","journal-title":"Med. Phys."},{"key":"bpexadcacabib4","doi-asserted-by":"publisher","first-page":"225","DOI":"10.1118\/1.2148918","article-title":"Iterative regularization in intensity-modulated radiation therapy optimization","volume":"33","author":"Carlsson","year":"2006","journal-title":"Med. Phys."},{"key":"bpexadcacabib5","doi-asserted-by":"publisher","first-page":"7648","DOI":"10.1002\/mp.15890","article-title":"Global optimization for spot-based treatment planning","volume":"49","author":"Chen","year":"2022","journal-title":"Med. Phys."},{"key":"bpexadcacabib6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3857\/roj.2018.00122","article-title":"Intensity-modulated radiation therapy: a review with a physics perspective","volume":"36","author":"Cho","year":"2018","journal-title":"Radiat Oncol J."},{"key":"bpexadcacabib7","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1002\/acm2.12224","article-title":"On the gEUD biological optimization objective for organs at risk in photon optimizer of eclipse treatment planning system","volume":"19","author":"Fogliata","year":"2017","journal-title":"J. Appl. Clin. Med. Phys."},{"key":"bpexadcacabib8","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1016\/j.ejmp.2021.05.003","article-title":"A review of deep learning based methods for medical image multi-organ segmentation","volume":"85","author":"Fu","year":"2021","journal-title":"Phys. Med."},{"key":"bpexadcacabib9","doi-asserted-by":"publisher","first-page":"617","DOI":"10.1088\/0031-9155\/52\/3\/006","article-title":"A finite size pencil beam algorithm for IMRT dose optimization: density corrections","volume":"52","author":"Jele\u0144","year":"2007","journal-title":"Phys. Med. Biol."},{"key":"bpexadcacabib10","doi-asserted-by":"publisher","first-page":"1747","DOI":"10.1088\/0031-9155\/50\/8\/009","article-title":"A finite size pencil beam for IMRT dose optimization","volume":"50","author":"Jele\u0144","year":"2005","journal-title":"Phys. Med. Biol."},{"key":"bpexadcacabib11","doi-asserted-by":"publisher","first-page":"2843","DOI":"10.1088\/0031-9155\/48\/17\/308","article-title":"Multiobjective inverse planning for intensity modulated radiotherapy with constraint-free gradient-based optimization algorithms","volume":"48","author":"Lahanas","year":"2003","journal-title":"Phys. Med. Biol."},{"key":"bpexadcacabib12","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.14558","article-title":"Clinical commissioning and introduction of an in-house artificial intelligence (AI) platform for automated head and neck intensity modulated radiation therapy (IMRT) treatment planning","volume":"26","author":"Li","year":"2025","journal-title":"J Appl Clin Med Phys"},{"key":"bpexadcacabib13","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/aba5eb","article-title":"Automatic IMRT planning via static field fluence prediction (AIP-SFFP): a deep learning algorithm for real-time prostate treatment planning","volume":"65","author":"Li","year":"2020","journal-title":"Phys. Med. Biol."},{"key":"bpexadcacabib14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13014-022-02045-y","article-title":"Applying pytorch toolkit to plan optimization for circular cone based robotic radiotherapy","volume":"17","author":"Liang","year":"2022","journal-title":"Radiation Oncology"},{"key":"bpexadcacabib15","doi-asserted-by":"publisher","first-page":"507","DOI":"10.1118\/1.2409749","article-title":"Reduction of IMRT beam complexity through the use of beam modulation penalties in the objective function","volume":"34","author":"Matuszak","year":"2007","journal-title":"Med. Phys."},{"key":"bpexadcacabib16","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1002\/acm2.13337","article-title":"Knowledge-based radiation treatment planning: a data-driven method survey","volume":"22","author":"Momin","year":"2021","journal-title":"J. Appl. Clin. Med. Phys."},{"key":"bpexadcacabib17","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1118\/1.598063","article-title":"Reporting and analyzing dose distributions: a concept of equivalent uniform dose","volume":"24","author":"Niemierko","year":"1997","journal-title":"Med. Phys."},{"key":"bpexadcacabib18","first-page":"8026","volume":"721","author":"Paszke","year":"2019"},{"key":"bpexadcacabib19","doi-asserted-by":"publisher","first-page":"111","DOI":"10.1016\/j.zemedi.2007.12.001","article-title":"A comparison of three optimization algorithms for intensity modulated radiation therapy","volume":"18","author":"Pflugfelder","year":"2008","journal-title":"Z. Med. Phys."},{"key":"bpexadcacabib20","doi-asserted-by":"publisher","first-page":"215","DOI":"10.1007\/s11750-008-0064-1","article-title":"Intensity modulated radiation therapy treatment plan optimization","volume":"16","author":"Romeijn","year":"2008","journal-title":"TOP"},{"key":"bpexadcacabib21","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2020.580919","article-title":"A review on application of deep learning algorithms in external beam radiotherapy automated treatment planning","volume":"10","author":"Wang","year":"2020","journal-title":"Front Oncol"},{"key":"bpexadcacabib22","doi-asserted-by":"publisher","first-page":"4611","DOI":"10.1088\/0031-9155\/49\/19\/011","article-title":"A dose calculation method including scatter for IMRT optimization","volume":"49","author":"Wu","year":"2004","journal-title":"Phys. Med. Biol."}],"container-title":["Biomedical Physics &amp; Engineering Express"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,22]],"date-time":"2025-04-22T13:49:37Z","timestamp":1745329777000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2057-1976\/adcaca"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,22]]},"references-count":22,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2025,4,22]]},"published-print":{"date-parts":[[2025,5,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2057-1976\/adcaca","relation":{},"ISSN":["2057-1976"],"issn-type":[{"value":"2057-1976","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4,22]]},"assertion":[{"value":"A machine learning toolkit assisted approach for IMRT fluence map optimization: feasibility and advantages","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 2025 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":"2025-03-18","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-04-09","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-04-22","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}