{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T12:16:16Z","timestamp":1789992976563,"version":"4.0.1"},"reference-count":53,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"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>Introduction<\/jats:title>\n                    <jats:p>Gliomas are the most common primary malignant tumors of the central nervous system, characterized by significant biological heterogeneity and poor prognosis. Multiparametric MRI-based image analysis has shown promise in prognostic assessment; however, effective integration of deep learning (DL), radiomics, and intratumoral heterogeneity (ITH) quantification remains underexplored. This study aims to develop an integrated prognostic model incorporating clinical and multiparametric MRI-derived features and systematically evaluate the predictive performance of different feature combinations.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This retrospective, multicenter study included 183 patients from the UCSF-TCIA cohort as the training set and 109 patients from Chongqing General Hospital as an independent test set. Based on preoperative multiparametric MRI\u2014including T1-weighted imaging (T1WI), T1-weighted contrast-enhanced imaging (T1CE), T2-weighted imaging (T2WI), T2 fluid-attenuated inversion recovery (T2-FLAIR), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC)\u2014radiomics, DL, and 2D\/3D ITH features were extracted. Feature selection was performed using Least Absolute Shrinkage and Selection Operator\u2013Cox (LASSO-Cox) regression, and both single and fusion models were constructed. Feature-level integration was achieved using a two-stage LASSO-Cox regression to leverage the complementary nature of information from different sources fully. Performance was quantified using the Harrell concordance index (C-index) and validated via Kaplan\u2013Meier survival analysis, time-dependent receiver operating characteristic (ROC) analysis, calibration curves, and decision curve analysis (DCA).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The fusion model integrating DL, radiomics, ITH, and clinical features (DenseNet121_DL+Rad+ITH) achieved C-indices of 0.801 (95% CI: 0.750\u20130.853) and 0.710 (95% CI: 0.638\u20130.783) on the training and test sets, respectively, outperforming all single models. The corresponding stacking ensemble yielded a comparable test C-index of 0.701. Kaplan\u2013Meier analysis confirmed significant risk stratification (P &amp;lt; 0.001). Time-dependent ROC analysis showed AUCs of 0.705, 0.721, 0.746, 0.800, and 0.873 at 12, 18, 24, 30, and 36 months. Calibration curves demonstrated good agreement between predicted and observed survival probabilities. DCA indicated favorable clinical net benefit for medium- to long-term survival prediction (24\u201336 months).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>An integrated prognostic model combining clinical characteristics with multiparametric MRI-derived features (DL, radiomics, and ITH) demonstrates strong prognostic potential for glioma patients. Fusion strategies generally outperform single-feature models; however, incremental gains from fusion are modest, and prospective validation in larger cohorts remains necessary.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fonc.2026.1873715","type":"journal-article","created":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T11:57:52Z","timestamp":1789991872000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["An integrated prognostic model incorporating clinical characteristics, multiparametric MRI radiomics, deep learning features, and intratumoral heterogeneity for glioma patients"],"prefix":"10.3389","volume":"16","author":[{"given":"Yanru","family":"Yang","sequence":"first","affiliation":[{"name":"School of Imaging Medicine, North Sichuan Medical College","place":["Nanchong, Sichuan, China"]},{"name":"Department of Radiology, Chongqing Academy of Medical Sciences, Chongqing General Hospital, Chongqing University","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Keqi","family":"Su","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chongqing Academy of Medical Sciences, Chongqing General Hospital, Chongqing University","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kaiqiang","family":"Tang","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chongqing Academy of Medical Sciences, Chongqing General Hospital, Chongqing University","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kang","family":"Li","sequence":"additional","affiliation":[{"name":"School of Imaging Medicine, North Sichuan Medical College","place":["Nanchong, Sichuan, China"]},{"name":"Department of Radiology, Chongqing Academy of Medical Sciences, Chongqing General Hospital, Chongqing University","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,21]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(23)01054-1","article-title":"Primary brain tumours in adults","volume":"402","author":"Van Den Bent","year":"2023","journal-title":"Lancet"},{"key":"B2","doi-asserted-by":"publisher","first-page":"iii1","DOI":"10.1093\/neuonc\/noab200","article-title":"CBTRUS statistical report: Primary brain and other central nervous system tumors diagnosed in the United States in 2014\u20132018","volume":"23","author":"Ostrom","year":"2021","journal-title":"Neuro-Oncology"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1002\/advs.201902971","article-title":"Advances in the knowledge of the molecular biology of glioblastoma and its impact in patient diagnosis, stratification, and treatment","volume":"7","author":"Delgado\u2010Mart\u00edn","year":"2020","journal-title":"Adv Sci"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.3390\/cancers18030491","article-title":"MRI-based radiomics for non-invasive prediction of molecular biomarkers in gliomas","volume":"18","author":"Agosti","year":"2026","journal-title":"Cancers"},{"key":"B5","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1186\/s40164-022-00277-y","article-title":"Heterogeneity of the tumor immune microenvironment and its clinical relevance","volume":"11","author":"Jia","year":"2022","journal-title":"Exp Hematol Oncol"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2024.02.041","article-title":"Cancer biomarkers: Emerging trends and clinical implications for personalized treatment","volume":"187","author":"Passaro","year":"2024","journal-title":"Cell"},{"key":"B7","doi-asserted-by":"publisher","first-page":"660","DOI":"10.1038\/s41746-025-02018-x","article-title":"GlioSurv: Interpretable transformer for multimodal, individualized survival prediction in diffuse glioma","volume":"8","author":"Lee","year":"2025","journal-title":"NPJ Digit Med"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1007\/s11547-023-01725-3","article-title":"Time-to-event overall survival prediction in glioblastoma multiforme patients using magnetic resonance imaging radiomics","volume":"128","author":"Hajianfar","year":"2023","journal-title":"Radiol Med"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1093\/neuonc\/noab106","article-title":"The 2021 WHO classification of tumors of the central nervous system: a summary","volume":"23","author":"Louis","year":"2021","journal-title":"Neuro-Oncol"},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1007\/s00432-021-03511-y","article-title":"A 25-year retrospective, single center analysis of 343 WHO grade II\/III glioma patients: Implications for grading and temozolomide therapy","volume":"147","author":"Steidl","year":"2021","journal-title":"J Cancer Res Clin Oncol"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.7717\/peerj.19906","article-title":"Survival prediction in gliomas based on MRI radiomics combined with clinical factors and molecular biomarkers","volume":"13","author":"Hao","year":"2025","journal-title":"PeerJ"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.7150\/ijbs.71046","article-title":"The progress of multimodal imaging combination and subregion based radiomics research of cancers","volume":"18","author":"Zhang","year":"2022","journal-title":"Int J Biol Sci"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.3390\/biomedicines11071957","article-title":"Assessment of hyperosmolar blood\u2013brain barrier opening in glioblastoma via histology with evans blue and DCE-MRI","volume":"11","author":"Conq","year":"2023","journal-title":"Biomedicines"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.3390\/cancers15112992","article-title":"Conventional and advanced magnetic resonance imaging assessment of non-enhancing peritumoral area in brain tumor","volume":"15","author":"Scola","year":"2023","journal-title":"Cancers"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.3390\/cancers15133267","article-title":"Recent advances in deep learning and medical imaging for head and neck cancer treatment: MRI, CT, and PET scans","volume":"15","author":"Illimoottil","year":"2023","journal-title":"Cancers"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.3390\/brainsci16050468","article-title":"Advancing brain tumor diagnosis using deep learning: A systematic and critical review on methodological approaches to glioma segmentation and classification through multiparametric MRI","volume":"16","author":"Aresta","year":"2026","journal-title":"Brain Sci"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1016\/j.jhepr.2022.100443","article-title":"Artificial intelligence in liver diseases: Improving diagnostics, prognostics and response prediction","volume":"4","author":"Nam","year":"2022","journal-title":"JHEP Rep"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejrad.2025.112406","article-title":"A novel multimodal framework combining habitat radiomics, deep learning, and conventional radiomics for predicting MGMT gene promoter methylation in glioma: Superior performance of integrated models","volume":"192","author":"Zhu","year":"2025","journal-title":"Eur J Radiol"},{"key":"B19","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1038\/s41698-025-00884-y","article-title":"MRI transformer deep learning and radiomics for predicting IDH wild type TERT promoter mutant gliomas","volume":"9","author":"Niu","year":"2025","journal-title":"NPJ Precis Onc"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.3390\/cancers17213417","article-title":"Radiomics and deep learning interplay for predicting MGMT methylation in glioblastoma: The crucial role of segmentation quality","volume":"17","author":"Lizzi","year":"2025","journal-title":"Cancers (Basel)"},{"key":"B21","doi-asserted-by":"publisher","first-page":"36756","DOI":"10.1038\/s41598-025-20711-5","article-title":"Integrating deep learning and radiomics for preoperative glioma grading using multi-center MRI data","volume":"15","author":"Yin","year":"2025","journal-title":"Sci Rep"},{"key":"B22","doi-asserted-by":"publisher","first-page":"893","DOI":"10.1007\/s00330-022-09055-0","article-title":"ITHscore: Comprehensive quantification of intra-tumor heterogeneity in NSCLC by multi-scale radiomic features","volume":"33","author":"Li","year":"2022","journal-title":"Eur Radio"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.1016\/j.acra.2024.04.008","article-title":"Quantification of intratumoral heterogeneity: Distinguishing histological subtypes in clinical T1 stage lung adenocarcinoma presenting as pure ground-glass nodules on computed tomography","volume":"31","author":"Zhang","year":"2024","journal-title":"Acad Radiol"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1007\/s11547-023-01676-9","article-title":"Radiomics under 2D regions, 3D regions, and peritumoral regions reveal tumor heterogeneity in non-small cell lung cancer: A multicenter study","volume":"128","author":"Zhang","year":"2023","journal-title":"Radiol Med"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.241805","article-title":"Nomogram for predicting neoadjuvant chemotherapy response in breast cancer using MRI-based intratumoral heterogeneity quantification","volume":"315","author":"Huang","year":"2025","journal-title":"Radiology"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.222830","article-title":"MRI-based quantification of intratumoral heterogeneity for predicting treatment response to neoadjuvant chemotherapy in breast cancer","volume":"308","author":"Shi","year":"2023","journal-title":"Radiology"},{"key":"B27","doi-asserted-by":"publisher","first-page":"42222","DOI":"10.1038\/s41598-025-26331-3","article-title":"CT-based intratumoral heterogeneity quantification fusing deep learning radiomics for predicting lymph node metastasis in early-stage lung adenocarcinoma: A multicenter study","volume":"15","author":"Hu","year":"2025","journal-title":"Sci Rep"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2025.109137","article-title":"Multiperspective tumor heterogeneity metrics for preoperative prediction of IASLC grading in clinical stage IA lung adenocarcinomas: A multicenter study","volume":"274","author":"Zuo","year":"2026","journal-title":"Comput Methods Programs BioMed"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.220058","article-title":"The university of california san francisco preoperative diffuse glioma MRI dataset","volume":"4","author":"Calabrese","year":"2022","journal-title":"Radiology: Artif Intell"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-013-9622-7","article-title":"The cancer imaging archive (TCIA): Maintaining and operating a public information repository","volume":"26","author":"Clark","year":"","journal-title":"J Digital Imaging"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0307825","article-title":"Precision meets generalization: Enhancing brain tumor classification via pretrained DenseNet with global average pooling and hyperparameter tuning","volume":"19","author":"Aziz","year":"2024","journal-title":"PloS One"},{"key":"B32","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1016\/j.csbj.2023.08.012","article-title":"Developing a weakly supervised deep learning framework for breast cancer diagnosis with HR status based on mammography images","volume":"22","author":"Zhang","year":"2023","journal-title":"Comput Struct Biotechnol J"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1158\/1078-0432.CCR-17-2236","article-title":"Residual convolutional neural network for the determination of IDH status in low- and high-grade gliomas from MR imaging","volume":"24","author":"Chang","year":"2018","journal-title":"Clin Cancer Res"},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.1093\/neuonc\/noab238","article-title":"Improving the noninvasive classification of glioma genetic subtype with deep learning and diffusion-weighted imaging","volume":"24","author":"Cluceru","year":"2022","journal-title":"Neuro-Oncol"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.3174\/ajnr.A6365","article-title":"Deep transfer learning and radiomics feature prediction of survival of patients with high-grade gliomas","volume":"41","author":"Han","year":"2020","journal-title":"Am J Neuroradiol"},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1186\/s13244-025-02097-0","article-title":"Intratumoral heterogeneity score enhances invasiveness prediction in pulmonary ground-glass nodules via stacking ensemble machine learning","volume":"16","author":"Zuo","year":"2025","journal-title":"Insights into Imaging"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.3389\/fimmu.2026.1898208","article-title":"CT-derived topological intratumoral heterogeneity predicts major pathological response to neoadjuvant immunochemotherapy in resectable non-small-cell lung cancer: A two-center study","volume":"17","author":"Zhou","year":"2026","journal-title":"Front Immunol"},{"key":"B38","doi-asserted-by":"publisher","first-page":"91","DOI":"10.1186\/s40164-026-00788-y","article-title":"Biology-informed risk stratification of glioblastoma by integrating MRI-based intratumoral heterogeneity with clinical features: A multicenter validation study","volume":"15","author":"Zhang","year":"2026","journal-title":"Exp Hematol Oncol"},{"key":"B39","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2026.1775958","article-title":"Radiomics-based models to predict IDH mutation status and prognosis in gliomas using MRI: A multicenter study","volume":"16","author":"S\u00fcmer-Arpak","year":"","journal-title":"Front Oncol"},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics16040512","article-title":"Machine learning-based prognosis prediction in glioblastoma multiforme patients by integrating clinical data with multimodal radiomics","volume":"16","author":"Huang","year":"2026","journal-title":"Diagnostics (Basel)"},{"key":"B41","doi-asserted-by":"publisher","DOI":"10.1007\/s11548-025-03548-1","article-title":"Glioblastoma survival prediction through MRI and clinical data integration with transfer learning","volume":"21","author":"Marasi","year":"2026","journal-title":"Int J Comput Assisted Radiol Surg"},{"key":"B42","doi-asserted-by":"publisher","DOI":"10.1016\/j.acra.2024.09.021","article-title":"Deep learning and habitat radiomics for the prediction of glioma pathology using multiparametric MRI: A multicenter study","volume":"32","author":"Zhu","year":"2025","journal-title":"Acad Radiol"},{"key":"B43","doi-asserted-by":"publisher","first-page":"77","DOI":"10.1186\/s13244-025-01950-6","article-title":"The clinical implications and interpretability of computational medical imaging (radiomics) in brain tumors","volume":"16","author":"Wang","year":"2025","journal-title":"Insights into Imaging"},{"key":"B44","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.trecan.2022.09.005","article-title":"Challenges in glioblastoma research: Focus on the tumor microenvironment","volume":"9","author":"Bikfalvi","year":"2023","journal-title":"Trends Cancer"},{"key":"B45","doi-asserted-by":"publisher","DOI":"10.1158\/2159-8290.CD-20-1474","article-title":"Diffuse glioma heterogeneity and its therapeutic implications","volume":"11","author":"Nicholson","year":"2021","journal-title":"Cancer Discov"},{"key":"B46","doi-asserted-by":"publisher","DOI":"10.1093\/neuonc\/noac166","article-title":"Combined molecular subtyping, grading, and segmentation of glioma using multi-task deep learning","volume":"25","author":"Van Der Voort","year":"2023","journal-title":"Neuro-Oncology"},{"key":"B47","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1186\/s40779-023-00458-8","article-title":"Artificial intelligence-driven radiomics study in cancer: The role of feature engineering and modeling","volume":"10","author":"Zhang","year":"2023","journal-title":"Mil Med Res"},{"key":"B48","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbae699","article-title":"Multimodal deep learning approaches for precision oncology: a comprehensive review","volume":"26","author":"Yang","year":"2024","journal-title":"Briefings Bioinf"},{"key":"B49","doi-asserted-by":"publisher","DOI":"10.1038\/s41416-021-01689-z","article-title":"High-dimensional role of AI and machine learning in cancer research","volume":"126","author":"Capobianco","year":"2022","journal-title":"Br J Cancer"},{"key":"B50","doi-asserted-by":"publisher","first-page":"385","DOI":"10.1186\/s12967-023-04222-3","article-title":"Assessment of brain cancer atlas maps with multimodal imaging features","volume":"21","author":"Capobianco","year":"2023","journal-title":"J Transl Med"},{"key":"B51","doi-asserted-by":"publisher","DOI":"10.3390\/cancers13040761","article-title":"Tumor heterogeneity in glioblastomas: From light microscopy to molecular pathology","volume":"13","author":"Becker","year":"2021","journal-title":"Cancers"},{"key":"B52","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1038\/s41582-025-01171-x","article-title":"Multiscale network perspectives on glioma: From tumour biology to symptoms, survival and treatment","volume":"22","author":"Douw","year":"2026","journal-title":"Nat Rev Neurol"},{"key":"B53","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2025.1743451","article-title":"Evaluating sequence contributions to MRI radiomics for glioblastoma survival: Single vs fusion models","volume":"15","author":"Guo","year":"","journal-title":"Front Oncol"}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1873715\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T11:57:53Z","timestamp":1789991873000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1873715\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,21]]},"references-count":53,"alternative-id":["10.3389\/fonc.2026.1873715"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2026.1873715","relation":{},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,21]]},"article-number":"1873715"}}