{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T05:07:53Z","timestamp":1784092073124,"version":"3.55.0"},"reference-count":35,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T00:00:00Z","timestamp":1784073600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Oncol."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Intensity-modulated radiation therapy (IMRT) is a treatment modality for delivering a higher radiation dose to the target using multiple radiation beams from various gantry angles. Since an IMRT plan is complex and presents a high risk, quality assurance is crucial. In routine clinical practice, a radiotherapy treatment plan review is manually conducted by experienced planners to ensure the high quality and reliability of an IMRT plan. To assist this procedure, several machine-learning methods have been developed to automatically identify outliers or anomalous plans from a large amount of treatment plans.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>Due to the highly complex and non-linear nature of most machine-learning methods, their results are less explainable, and therefore, cannot be fully trusted by clinical users. To alleviate this issue, an unsupervised deep-learning model with an interpretable tool was introduced and evaluated on our clinical database.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Six hundred IMRT treatment plans were collected from our institute and the relevant features were extracted. A standard autoencoder (AE) was used to build an anomaly detection model from the normal plans and identify anomalies with exceptionally high reconstruction errors. To help explain the model output, a feature perturbation interpretation (FPI) method was introduced to rank the impacts of the features on the detected anomalies. To validate the effectiveness of the FPI method, the feature ranks provided by the local depth-based isolation forest feature importance (Local-DIFFI) and SHapley Additive exPlanations (SHAP) methods were compared. In addition, the effectiveness of the FPI method was evaluated on the other four classic detection models, local discrete factor (LOF), density-based spatial clustering of applications with noise (HDBSCAN), one class of support vector machines (OC-SVM), and principal component analysis (PCA).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The AE achieved the best detection performance among all the methods. The area under the curve (AUC) value of the AE was 0.98, and the average accuracy, precision, and F1 score were 0.91, 0.61, and 0.74, respectively. The top five features with the largest impacts on the detected anomalies obtained by FPI, local-DIFFI, and SHAP are the same.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The AE is an effective deep-learning model in identifying anomalous plans compared to classic detection models. The FPI method is universal and a reliable tool in determining the feature impact on the anomalies. Their combination provided a highly interpretable deep learning model for automatic plan review in radiotherapy.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fonc.2026.1823869","type":"journal-article","created":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T04:47:23Z","timestamp":1784090843000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Evaluation of an interpretable deep-learning model for the automated plan review of intensity-modulated radiation therapy"],"prefix":"10.3389","volume":"16","author":[{"given":"Yuhan","family":"Fan","sequence":"first","affiliation":[{"name":"Department of Radiation Oncology, National Cancer Center\/National Clinical Research Center for Cancer\/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiawen","family":"Shang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Cardiovascular Disease, Fuwai Hospital, National Center for Cardiovascular Disease, Chinese Academy of Medical Sciences and Peking Union Medical College","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ke","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology, National Cancer Center\/National Clinical Research Center for Cancer\/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhihui","family":"Hu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology, National Cancer Center\/National Clinical Research Center for Cancer\/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiqiang","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology, National Cancer Center\/National Clinical Research Center for Cancer\/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hui","family":"Yan","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology, National Cancer Center\/National Clinical Research Center for Cancer\/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peng","family":"Huang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology, National Cancer Center\/National Clinical Research Center for Cancer\/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"81","DOI":"10.11910\/j.issn.2791-3937.2022.20220006","article-title":"Modern radiotherapy in the multidisciplinary management of common cancers","volume":"1","author":"Du","year":"2022","journal-title":"Clin Cancer Bull"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1002\/1878-0261.12712","article-title":"Role of radiation oncology in modern multidisciplinary cancer treatment","volume":"14","author":"Valentini","year":"2020","journal-title":"Mol Oncol"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1046\/j.1440-1622.2002.02328.x","article-title":"Modern radiotherapy for modern surgeons: an update on radiation oncology","volume":"72","author":"Kenny","year":"2002","journal-title":"Anz J Surg"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.hoc.2019.08.005","article-title":"Modern radiation therapy planning and delivery","volume":"33","author":"Gardner","year":"2019","journal-title":"Hematol Oncol Clin North Am"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1016\/j.hoc.2019.08.008","article-title":"Imaging for target delineation and treatment planning in radiation oncology: current and emerging techniques","volume":"33","author":"Stieb","year":"2019","journal-title":"Hematol Oncol Clin North Am"},{"key":"B6","doi-asserted-by":"publisher","first-page":"45","DOI":"10.1016\/j.hoc.2006.01.006","article-title":"Modern radiation treatment planning and delivery--from r\u00f6ntgen to real time","volume":"20","author":"Smith","year":"2006","journal-title":"Hematol Oncol Clin North Am"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1118\/1.4736825","article-title":"Technical note: electronic chart checks in a paperless radiation therapy clinic","volume":"39","author":"Yang","year":"2012","journal-title":"Med Phys"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2007.05.092","article-title":"Quality assurance of intensity-modulated radiation therapy","volume":"71","author":"Palta","year":"2008","journal-title":"Int J Radiat Oncol Biol Phys"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1118\/1.4961010","article-title":"The effectiveness of pretreatment physics plan review for detecting errors in radiation therapy","volume":"43","author":"Gopan","year":"2016","journal-title":"Med Phys"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1002\/acm2.12845","article-title":"Efficiency and safety increases after the implementation of a multi-institutional automated plan check tool at our institution","volume":"21","author":"Berry","year":"2020","journal-title":"J Appl Clin Med Phys"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1120\/jacmp.v16i3.5396","article-title":"AutoLock: a semiautomated system for radiotherapy treatment plan quality control","volume":"16","author":"Dewhurst","year":"2015","journal-title":"J Appl Clin Med Phys"},{"key":"B12","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1120\/jacmp.v17i6.6322","article-title":"Improving treatment plan evaluation with automation","volume":"17","author":"Covington","year":"2016","journal-title":"J Appl Clin Med Phys"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1120\/jacmp.v10i1.2855","article-title":"Automating the initial physics chart\u2010checking process","volume":"10","author":"Furhang","year":"2009","journal-title":"J Appl Clin Med Phys"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1118\/1.3683646","article-title":"Automated radiotherapy treatment plan integrity verification","volume":"39","author":"Yang","year":"2012","journal-title":"Med Phys"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2021.102260","article-title":"Population-based 3D respiratory motion modelling from convolutional autoencoders for 2D ultrasound-guided radiotherapy","volume":"75","author":"Mezheritsky","year":"2022","journal-title":"Med Image Anal"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1016\/j.artmed.2022.102387","article-title":"A deep LSTM autoencoder-based framework for predictive maintenance of a proton radiotherapy delivery system","volume":"132","author":"Dou","year":"2022","journal-title":"Artif Intell Med"},{"key":"B17","doi-asserted-by":"publisher","first-page":"235023","DOI":"10.1088\/1361-6560\/abb31c","article-title":"Multi-task autoencoder based classification-regression model for patient-specific VMAT QA","volume":"65","author":"Wang","year":"2020","journal-title":"Phys Med Biol"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1002\/mp.16210","article-title":"A synthesized gamma distribution-based patient-specific VMAT QA using a generative adversarial network","volume":"50","author":"Matsuura","year":"2023","journal-title":"Med Phys"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1016\/j.radonc.2021.06.024","article-title":"Commissioning and clinical implementation of an autoencoder based classification-regression model for VMAT patient-specific QA in a multi-institution scenario","volume":"161","author":"Yang","year":"2021","journal-title":"Radiother Oncol"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.3390\/bioengineering10040405","article-title":"Anomaly detection for sensor signals utilizing deep learning autoencoder-based neural networks","volume":"10","author":"Esmaeili","year":"2023","journal-title":"Bioeng (Basel)"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-021-00558-8","article-title":"Unsupervised deep anomaly detection for medical images using an improved adversarial autoencoder","volume":"35","author":"Zhang","year":"2022","journal-title":"J Digit Imaging"},{"key":"B22","doi-asserted-by":"publisher","first-page":"216","DOI":"10.21037\/tcr-2025-1-2641","article-title":"Artificial intelligence applications in adaptive radiotherapy-a narrative review","volume":"15","author":"Chen","year":"2026","journal-title":"Transl Cancer Res"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1016\/s2589-7500(21)00208-9","article-title":"The false hope of current approaches to explainable artificial intelligence in health care","volume":"3","author":"Ghassemi","year":"2021","journal-title":"Lancet Digit Health"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102470","article-title":"Explainable artificial intelligence (XAI) in deep learning-based medical image analysis","volume":"79","author":"van der Velden","year":"2022","journal-title":"Med Image Anal"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.57728","article-title":"AI-driven clinical decision support systems: an ongoing pursuit of potential","volume":"16","author":"Elhaddad","year":"2024","journal-title":"Cureus"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.3390\/jimaging8090229","article-title":"Shapley-additive-explanations-based factor analysis for dengue severity prediction using machine learning","volume":"8","author":"Chowdhury","year":"2022","journal-title":"J Imaging"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2024.109569","article-title":"Local interpretable model-agnostic explanation approach for medical imaging analysis: a systematic literature review","volume":"185","author":"Hassan","year":"2025","journal-title":"Comput Biol Med"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1109\/access.2024.3368377","article-title":"Explainable AI for intrusion detection systems: LIME and SHAP applicability on multi-layer perceptron","volume":"12","author":"Gaspar","year":"2024","journal-title":"IEEE Access"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1016\/j.engappai.2022.105730","article-title":"Interpretable anomaly detection with DIFFI: depth-based feature importance of isolation forest","volume":"119","author":"Carletti","year":"2023","journal-title":"Eng Appl Artif Intell"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.1016\/j.ymssp.2021.108105","article-title":"An explainable artificial intelligence approach for unsupervised fault detection and diagnosis in rotating machinery","volume":"163","author":"Brito","year":"2022","journal-title":"Mech Syst Signal Process"},{"key":"B31","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1145\/335191.335388","article-title":"LOF: identifying density-based local outliers","volume":"29","author":"Breunig","year":"2000","journal-title":"ACM SIGMOD Rec"},{"key":"B32","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-642-37456-2_14","article-title":"Density-based clustering based on hierarchical density estimates","volume-title":"Advances in Knowledge Discovery and Data Mining","author":"Campello","year":"2013"},{"key":"B33","article-title":"Support vector method for novelty detection","volume-title":"Proceedings of the 12th International Conference on Neural Information Processing Systems","author":"Sch\u00f6lkopf","year":"1999"},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.1080\/14786440109462720","article-title":"LIII. On lines and planes of closest fit to systems of points in space","volume":"2","author":"Pearson","year":"1901","journal-title":"Lond Edinb Dublin Philos Mag J Sci"},{"key":"B35","doi-asserted-by":"publisher","first-page":"6171","DOI":"10.1038\/s41598-026-36548-5","article-title":"Anomaly detection of hyperspectral images based on improved isolation forest algorithm","volume":"16","author":"Li","year":"2026","journal-title":"Sci Rep"}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1823869\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T04:47:32Z","timestamp":1784090852000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1823869\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,15]]},"references-count":35,"alternative-id":["10.3389\/fonc.2026.1823869"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2026.1823869","relation":{},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,15]]},"article-number":"1823869"}}