{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T20:32:03Z","timestamp":1786393923376,"version":"build-2736575974"},"reference-count":59,"publisher":"Wiley","issue":"7","license":[{"start":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T00:00:00Z","timestamp":1782777600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"},{"start":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T00:00:00Z","timestamp":1782777600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/100011676","name":"Elekta","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100011676","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada","doi-asserted-by":"publisher","award":["RGPIN\u20102019\u201004715"],"award-info":[{"award-number":["RGPIN\u20102019\u201004715"]}],"id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Inverse planning is often used for Gamma Knife radiosurgery, allowing clinicians to mathematically specify desired clinical objectives and dose limits. The objectives are controlled by weights that are manually tuned to find the desired trade\u2010off, which varies from case to case. Automation of this process can reduce clinical workload and improve consistency in plan quality.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To train a deep reinforcement learning agent using a reward function that incorporates the clinical metrics from past plans into its scoring criteria. The metric trade\u2010off from the clinical plan is scored higher than all others, guiding the agent to produce plans with similar trade\u2010offs.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>An agent was trained to adjust the two priority weights (i.e., digital slider bars) in the clinical inverse planner. The agent consists of a neural network that receives the metrics and dose distribution of the current plan and the target and organ\u2010at\u2010risk masks as inputs. These methods were demonstrated on a dataset of 204 single\u2010target metastases and a dataset of 71 acoustic neuroma cases. The cases were split into training, validation, and testing sets of size 123\/41\/40 and 42\/14\/15 for the metastases and acoustic neuromas, respectively.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      On the metastases test dataset, the agent achieved a significantly higher (\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0=\u00a00.0136) average plan score (3.925\u00a0\u00b1\u00a00.130) compared to the default slider plans (3.874\u00a0\u00b1\u00a00.147). On the acoustic neuromas test dataset, the agent achieved a higher (\n                      <jats:italic>p<\/jats:italic>\n                      \u00a0=\u00a00.4493) average plan score (4.035\u00a0\u00b1\u00a00.177) compared to the default slider plans (3.995\u00a0\u00b1\u00a00.365). The higher plan scores are reflected in the four plan quality metrics: the agent's plans, on average, had metrics more similar to the clinical plans, compared to the default slider plans, for both test datasets.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed reward function enabled the agent to learn to find plans that aligned with historical planning decisions. Future work will investigate providing the agent with additional inputs that can explain the variability in planning decisions, which would further improve its performance.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70546","type":"journal-article","created":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T13:28:09Z","timestamp":1782826089000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Gamma Knife treatment planning using knowledge\u2010based reinforcement learning"],"prefix":"10.1002","volume":"53","author":[{"given":"Christopher","family":"Huynh","sequence":"first","affiliation":[{"name":"Medical Physics Department Sunnybrook Health Sciences Centre  Toronto Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bj\u00f6rn","family":"Ahlgren","sequence":"additional","affiliation":[{"name":"Elekta Instrument AB  Stockholm Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Beibei","family":"Zhang","sequence":"additional","affiliation":[{"name":"Medical Physics Department Sunnybrook Health Sciences Centre  Toronto Canada"},{"name":"Department of Radiation Oncology University of Toronto  Toronto Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dominik","family":"Fay","sequence":"additional","affiliation":[{"name":"Elekta Instrument AB  Stockholm Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"H\u00e5kan","family":"Nordstr\u00f6m","sequence":"additional","affiliation":[{"name":"Elekta Instrument AB  Stockholm Sweden"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mark","family":"Ruschin","sequence":"additional","affiliation":[{"name":"Medical Physics Department Sunnybrook Health Sciences Centre  Toronto Canada"},{"name":"Department of Radiation Oncology University of Toronto  Toronto Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,6,30]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.3171\/2014.7.GKS141199"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13440"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13337"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3253464"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.3539749"},{"key":"e_1_2_8_7_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4761864"},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4757927"},{"key":"e_1_2_8_9_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4769424"},{"key":"e_1_2_8_10_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4828788"},{"key":"e_1_2_8_11_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4906183"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1118\/1.4938583"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13014\u2010017\u20100806\u2010z"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/aa71f8"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2019.00750"},{"key":"e_1_2_8_16_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010019\u201052262\u2010x"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.adro.2021.100672"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14305"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13262"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13271"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010018\u201037741\u2010x"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ab039b"},{"key":"e_1_2_8_23_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13490"},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13597"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.13896"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12849"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41598\u2010020\u201068062\u20107"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2023.102533"},{"key":"e_1_2_8_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17058"},{"key":"e_1_2_8_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2020.03.016"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14236"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature24270"},{"key":"e_1_2_8_33_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.aar6404"},{"key":"e_1_2_8_34_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586\u2010019\u20101724\u2010z"},{"key":"e_1_2_8_35_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2024.104498"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ab18bf"},{"key":"e_1_2_8_37_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14114"},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14712"},{"key":"e_1_2_8_39_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ac09a2"},{"key":"e_1_2_8_40_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15576"},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2023.109685"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.14544"},{"key":"e_1_2_8_43_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361\u20106560\/ac2bb5"},{"key":"e_1_2_8_44_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejmp.2021.12.009"},{"key":"e_1_2_8_45_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.17100"},{"key":"e_1_2_8_46_1","unstructured":"MnihV KavukcuogluK SilverD et\u00a0al.Playing Atari with Deep Reinforcement Learning. Published online 2013.https:\/\/arxiv.org\/abs\/1312.5602"},{"key":"e_1_2_8_47_1","doi-asserted-by":"publisher","DOI":"10.1561\/2200000071"},{"key":"e_1_2_8_48_1","unstructured":"KingmaDP AdamBaJ:A Method for Stochastic Optimization. Published online2017.https:\/\/arxiv.org\/abs\/1412.6980"},{"key":"e_1_2_8_49_1","unstructured":"LillicrapTP HuntJJ PritzelA et\u00a0al.Continuous control with deep reinforcement learning. In:International Conference on Learning Representations. ICLR;2016."},{"key":"e_1_2_8_50_1","unstructured":"CardosoMJ LiW BrownR et\u00a0al.MONAI: an open\u2010source framework for deep learning in healthcare. Published online 2022.https:\/\/arxiv.org\/abs\/2211.02701"},{"key":"e_1_2_8_51_1","unstructured":"TanM LeQV.EfficientNet: rethinking model scaling for convolutional neural networks. Published online 2020."},{"key":"e_1_2_8_52_1","doi-asserted-by":"crossref","unstructured":"SandlerM HowardA ZhuM ZhmoginovA ChenLC.MobileNetV2: inverted residuals and linear bottlenecks. In:2018IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE Computer Society;2018:4510\u20104520. doi:10.1109\/CVPR.2018.00474","DOI":"10.1109\/CVPR.2018.00474"},{"key":"e_1_2_8_53_1","doi-asserted-by":"crossref","unstructured":"TanM ChenB PangR et\u00a0al.MnasNet: platform\u2010aware neural architecture search for mobile. In:2019IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE Computer Society;2019:2815\u20102823. doi:10.1109\/CVPR.2019.00293","DOI":"10.1109\/CVPR.2019.00293"},{"key":"e_1_2_8_54_1","doi-asserted-by":"crossref","unstructured":"HuJ ShenL SunG.Squeeze\u2010and\u2010Excitation Networks. In:2018IEEE\/CVF Conference on Computer Vision and Pattern Recognition.2018:7132\u20107141. doi:10.1109\/CVPR.2018.00745","DOI":"10.1109\/CVPR.2018.00745"},{"key":"e_1_2_8_55_1","doi-asserted-by":"crossref","unstructured":"HowardA SandlerM ChenB et\u00a0al.Searching for MobileNetV3. In:2019IEEE\/CVF International Conference on Computer Vision (ICCV).2019:1314\u20101324. doi:10.1109\/ICCV.2019.00140","DOI":"10.1109\/ICCV.2019.00140"},{"issue":"56","key":"e_1_2_8_56_1","first-page":"1929","article-title":"A simple way to prevent neural networks from overfitting","volume":"15","author":"Srivastava N","year":"2014","journal-title":"J Mach Learn Res"},{"key":"e_1_2_8_57_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_39"},{"key":"e_1_2_8_58_1","unstructured":"IoffeS SzegedyC.Batch normalization: accelerating deep network training by reducing internal covariate shift. In:BachF BleiD eds.Proceedings of the 32nd International Conference on Machine Learning. Vol37. Proceedings of Machine Learning Research. PMLR; 2015:448\u2010456.https:\/\/proceedings.mlr.press\/v37\/ioffe15.html"},{"key":"e_1_2_8_59_1","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.13936"},{"key":"e_1_2_8_60_1","doi-asserted-by":"publisher","DOI":"10.1109\/TRPMS.2021.3107454"}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70546","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70546","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70546","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T19:54:31Z","timestamp":1786391671000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70546"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,30]]},"references-count":59,"journal-issue":{"issue":"7","published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1002\/mp.70546"],"URL":"https:\/\/doi.org\/10.1002\/mp.70546","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,30]]},"assertion":[{"value":"2025-11-24","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-20","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-06-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70546"}}