{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,23]],"date-time":"2026-02-23T13:02:36Z","timestamp":1771851756209,"version":"3.50.1"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,6,6]],"date-time":"2025-06-06T00:00:00Z","timestamp":1749168000000},"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><jats:title>Purpose<\/jats:title><jats:p>Accurate differentiation between glioma recurrence and radiation necrosis is critical for the management of patients suspected of glioma recurrence following radiation therapy. This study aims to develop a deep learning-based methodology for automated discrimination between glioma recurrence and radiation necrosis using routine magnetic resonance imaging (MRI) scans.<\/jats:p><\/jats:sec><jats:sec><jats:title>Method<\/jats:title><jats:p>We retrospectively investigated 234 patients who underwent radiotherapy after glioma resection and presented with suspected recurrent lesions during follow-up MRI examinations. Routine 3D-MRI scans, including T1-weighted, T2-weighted, and contrast-enhanced T1 (T1ce) sequences, were acquired for each patient. Among the analyzed cases, 192 (82.1%) were pathologically confirmed as glioma recurrence, while 42 (17.9%) were diagnosed as radiation necrosis. Various Convolutional Neural Network (CNN) models were employed to learn radiological features indicative of glioma recurrence and radiation necrosis from the MRI scans. Performance evaluation metrics, such as sensitivity, specificity, accuracy, and area under the curve (AUC), were used to assess the models\u2019 performance.<\/jats:p><\/jats:sec><jats:sec><jats:title>Result<\/jats:title><jats:p>Among the evaluated CNN models, ResNet10 demonstrated the highest sensitivity (0.78), specificity (0.94), accuracy (0.91), and an AUC value of 0.83. Additionally, the MresNet model achieved the highest specificity (0.980) but exhibited a relatively lower sensitivity (0.56). Another evaluated CNN model, Vgg16, showed a sensitivity of 0.56, specificity of 0.94, accuracy of 0.88, and an AUC value of 0.70.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>The proposed ResNet10 CNN model demonstrates promising performance on routine MRI scans, rendering it highly applicable in clinical settings. These findings contribute to enhancing the diagnostic accuracy for distinguishing between glioma recurrence and radiation necrosis using routine MRI.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fonc.2025.1573700","type":"journal-article","created":{"date-parts":[[2025,6,6]],"date-time":"2025-06-06T04:10:14Z","timestamp":1749183014000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Development and validation of a deep learning algorithm for discriminating glioma recurrence from radiation necrosis on MRI"],"prefix":"10.3389","volume":"15","author":[{"given":"Yu-Zhe","family":"Ying","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiao-Hong","family":"Cai","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Han","family":"Yang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hua-Wei","family":"Huang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dao","family":"Zheng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao-Yi","family":"Li","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ge-Hong","family":"Dong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yong-Gang","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhong-Li","family":"Jiang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhu-Lin","family":"An","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guo-Bin","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,6,6]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"896","DOI":"10.1093\/neuonc\/nou087","article-title":"The epidemiology of glioma in adults: a \u201cstate of the science\u201d review","volume":"16","author":"Ostrom","year":"2014","journal-title":"Neuro Oncol"},{"key":"B2","article-title":"Organoid models of glioblastoma: advances, applications and challenges","volume":"10","author":"Zhang","year":"2020","journal-title":"Am J Cancer Res"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1056\/NEJMoa043330","article-title":"Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma","volume":"352","author":"Stupp","year":"2005","journal-title":"N Engl J Med"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.3390\/ijms150711832","article-title":"The diagnosis and treatment of pseudoprogression, radiation necrosis and brain tumor recurrence","volume":"15","author":"Parvez","year":"2014","journal-title":"Int J Mol Sci"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1093\/neuonc\/nos307","article-title":"Differentiating tumor recurrence from treatment necrosis: a review of neurooncologic imaging strategies","volume":"15","author":"Verma","year":"2013","journal-title":"Neuro Oncol"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s11060-009-9897-1","article-title":"Glioma recurrence versus radiation necrosis: accuracy of current imaging modalities","volume":"95","author":"Alexiou","year":"2009","journal-title":"J Neurooncol"},{"key":"B7","doi-asserted-by":"publisher","first-page":"11","DOI":"10.1111\/ajco.2014.10.issue-1","article-title":"Cerebral radiation necrosis","volume":"10","author":"Na","year":"2014","journal-title":"Asia Pac J Clin Oncol"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1038\/nm.2268","article-title":"Differentiation between glioma and radiation necrosis using molecular magnetic resonance imaging of endogenous proteins and peptides","volume":"17","author":"Zhou","year":"2011","journal-title":"Nat Med"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.3174\/ajnr.A1362","article-title":"Distinguishing recurrent intra-axial metastatic tumor from radiation necrosis following gamma knife radiosurgery using dynamic susceptibility-weighted contrast-enhanced perfusion MR imaging","volume":"30","author":"Barajas","year":"2009","journal-title":"AJNR Am J Neuroradiol"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.3174\/ajnr.A1580","article-title":"MR spectroscopy in radiation injury","volume":"30","author":"Sundgren","year":"2009","journal-title":"AJNR Am J Neuroradiol"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1007\/s00234-010-0731-4","article-title":"Distinction between postoperative recurrent glioma and radiation injury using MR diffusion tensor imaging","volume":"52","author":"Xu","year":"2010","journal-title":"Neuroradiology"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.18632\/oncotarget.19024","article-title":"The performance of 11C-Methionine PET in the differential diagnosis of glioma recurrence","volume":"8","author":"Xu","year":"2017","journal-title":"Oncotarget"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.2176\/nmc.oa2013-0117","article-title":"Comparison of (11)C-methionine, (11)C-choline, and (18)F-fluorodeoxyglucose-PET for distinguishing glioma recurrence from radiation necrosis","volume":"54","author":"Takenaka","year":"2014","journal-title":"Neurol Med Chir (Tokyo)"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1155\/2019\/2893043","article-title":"Differentiation of recurrence from radiation necrosis in gliomas based on the radiomics of combinational features and multimodality MRI images","volume":"2019","author":"Zhang","year":"2019","journal-title":"Comput Math Methods Med"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-017-5154-8","article-title":"A predictive model for distinguishing radiation necrosis from tumour progression after gamma knife radiosurgery based on radiomic features from MR images","volume":"28","author":"Zhang","year":"2018","journal-title":"Eur Radiol"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.3174\/ajnr.A4931","article-title":"Computer-extracted texture features to distinguish cerebral radionecrosis from recurrent brain tumors on multiparametric MRI: A feasibility study","volume":"37","author":"Tiwari","year":"2016","journal-title":"AJNR Am J Neuroradiol"},{"key":"B17","doi-asserted-by":"publisher","first-page":"e19805","DOI":"10.2196\/19805","article-title":"Deep learning methodology for differentiating glioma recurrence from radiation necrosis using multimodal magnetic resonance imaging: algorithm development and validation","volume":"8","author":"Gao","year":"2020","journal-title":"JMIR Med Inform"},{"key":"B18","doi-asserted-by":"publisher","first-page":"vdae199","DOI":"10.1093\/noajnl\/vdae199","article-title":"Deep learning-based postoperative glioblastoma segmentation and extent of resection evaluation: Development, external validation, and model comparison","volume":"6","author":"Cepeda","year":"2024","journal-title":"Neurooncol Adv"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1053\/j.semnuclmed.2020.08.002","article-title":"Intelligent imaging in nuclear medicine: the principles of artificial intelligence, machine learning and deep learning","volume":"51","author":"Currie","year":"2021","journal-title":"Semin Nucl Med"},{"key":"B20","doi-asserted-by":"publisher","first-page":"101727","DOI":"10.1016\/j.nicl.2019.101727","article-title":"Inter-rater agreement in glioma segmentations on longitudinal MRI","volume":"22","author":"Visser","year":"2019","journal-title":"NeuroImage Clin"},{"key":"B21","doi-asserted-by":"publisher","first-page":"26","DOI":"10.1186\/s40708-023-00207-6","article-title":"Deep learning-based algorithm for postoperative glioblastoma MRI segmentation: a promising new tool for tumor burden assessment","volume":"10","author":"Bianconi","year":"2023","journal-title":"Brain Inform"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1878","DOI":"10.3390\/biomedicines12081878","article-title":"Deep learning for MRI segmentation and molecular subtyping in glioblastoma: critical aspects from an emerging field","volume":"12","author":"Bonada","year":"2024","journal-title":"Biomedicines"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.3389\/fcell.2021.686894","article-title":"DeepOMe: A web server for the prediction of 2\u2019-O-me sites based on the hybrid CNN and BLSTM architecture","volume":"9","author":"Li","year":"2021","journal-title":"Front Cell Dev Biol"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1155\/2018\/6828396","article-title":"Radiation necrosis, pseudoprogression, pseudoresponse, and tumor recurrence: imaging challenges for the evaluation of treated gliomas","volume":"2018","author":"Zikou","year":"2018","journal-title":"Contrast Media Mol Imaging"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.3174\/ajnr.A2397","article-title":"Pseudoprogression and pseudoresponse: imaging challenges in the assessment of posttreatment glioma","volume":"32","author":"Hygino da Cruz","year":"2011","journal-title":"AJNR Am J Neuroradiol"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2017161588","article-title":"Differentiation of recurrent glioblastoma from delayed radiation necrosis by using voxel-based multiparametric analysis of MR imaging data","volume":"285","author":"Yoon","year":"2017","journal-title":"Radiology"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1148\/radiology.217.2.r00nv36377","article-title":"Malignant gliomas: MR imaging spectrum of radiation therapy- and chemotherapy-induced necrosis of the brain after treatment","volume":"217","author":"Kumar","year":"2000","journal-title":"Radiology"},{"key":"B28","doi-asserted-by":"publisher","first-page":"e11556","DOI":"10.1097\/MD.0000000000011556","article-title":"Accuracy of 11C-choline positron emission tomography in differentiating glioma recurrence from radiation necrosis: A systematic review and meta-analysis","volume":"97","author":"Gao","year":"2018","journal-title":"Med (Baltimore)"},{"key":"B29","doi-asserted-by":"publisher","first-page":"e0141438","DOI":"10.1371\/journal.pone.0141438","article-title":"Differentiating radiation-induced necrosis from recurrent brain tumor using MR perfusion and spectroscopy: A meta-analysis","volume":"11","author":"Chuang","year":"2016","journal-title":"PloS One"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.1007\/s11060-011-0694-2","article-title":"Glioma residual or recurrence versus radiation necrosis: accuracy of pentavalent technetium-99m-dimercaptosuccinic acid [Tc-99m (V) DMSA] brain SPECT compared to proton magnetic resonance spectroscopy (1H-MRS): initial results","volume":"106","author":"Amin","year":"2012","journal-title":"J Neurooncol"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1093\/neuonc\/noab106","article-title":"The\u00a02021 WHO classification of tumors of the central nervous system: a summary","volume":"23","author":"Louis","year":"2021","journal-title":"Neuro Oncol"},{"key":"B32","doi-asserted-by":"publisher","first-page":"188721","DOI":"10.1016\/j.bbcan.2022.188721","article-title":"Proteomics and metabolomics approach in adult and pediatric glioma diagnostics","volume":"1877","author":"Pienkowski","year":"2022","journal-title":"Biochim Biophys Acta Rev Cancer"}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2025.1573700\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,6]],"date-time":"2025-06-06T04:10:14Z","timestamp":1749183014000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2025.1573700\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,6]]},"references-count":32,"alternative-id":["10.3389\/fonc.2025.1573700"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2025.1573700","relation":{},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,6]]},"article-number":"1573700"}}