{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T06:12:24Z","timestamp":1790143944481,"version":"4.0.1"},"reference-count":57,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T00:00:00Z","timestamp":1790121600000},"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. Radiol."],"abstract":"<jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>\n                      The purpose of this study was to evaluate synthetic T\n                      <jats:sub>1<\/jats:sub>\n                      weighted post-contrast MR images generated from non-contrast-enhanced MR images, using deep learning (DL) methods, for automated brain tumor segmentation and automated classification of glioma grade based on imaging characteristics.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>\n                      Three DL models were developed for synthesizing T\n                      <jats:sub>1<\/jats:sub>\n                      -weighted post-contrast MRI (T\n                      <jats:sub>1<\/jats:sub>\n                      ce) images using a publicly available dataset: (1) a conditional neural field with shift modulation (CoNeS) model, (2) a denoising diffusion probabilistic model (DDPM), and (3) a hybrid CoNeS\u2009+\u2009DDPM model. Five experimental settings were evaluated, incorporating combinations of real or synthetic T\n                      <jats:sub>1<\/jats:sub>\n                      ce and pre-contrast images. Synthetic T\n                      <jats:sub>1<\/jats:sub>\n                      ce images were evaluated in terms of image quality metrics, accuracy of tumor segmentation, and classification of cases into high-grade vs. low-grade glioma. Two classifier families were evaluated: (a) radiomics-based machine learning classifiers using random forest on extracted radiomic features, and (b) deep learning classifiers employing a convolutional neural network.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The combined CoNeS\u2009+\u2009DDPM model performed best in image quality and similarity metrics on the test set. Bootstrap analysis revealed that radiomics-based ML and deep learning-based classifiers exhibited distinct characteristics, each of which outperforming the other in different metrics.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>\n                      The present study demonstrates significant advances in medical image AI synthesis by integrating stable diffusion and conditional neural fields, improving the overall quality of synthetic T\n                      <jats:sub>1<\/jats:sub>\n                      ce images. Nonetheless, synthetic images generated solely from pre-contrast sequences failed to consistently reproduce clinically relevant glioma features. As a result, current AI-generated T\n                      <jats:sub>1<\/jats:sub>\n                      ce images still lack the diagnostic fidelity required to replace true contrast-enhanced imaging in clinical practice.\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fradi.2026.1931738","type":"journal-article","created":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T05:49:34Z","timestamp":1790142574000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Downstream evaluation of synthetic AI-generated T1-weighted contrast-enhanced MR images in glioma segmentation and grading"],"prefix":"10.3389","volume":"6","author":[{"given":"Dimosthenis E.","family":"Gkotsis","sequence":"first","affiliation":[{"name":"Pharmaceutical Diagnostics, GE HealthCare","place":["Athens, Greece"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Adam J.","family":"Schwarz","sequence":"additional","affiliation":[{"name":"Pharmaceutical Diagnostics, GE HealthCare","place":["Marlborough, MA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jan","family":"Wolber","sequence":"additional","affiliation":[{"name":"Pharmaceutical Diagnostics, GE HealthCare, Chalfont Saint Giles","place":["United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lehel M.","family":"Ferenczi","sequence":"additional","affiliation":[{"name":"Science and Technology Organization, GE HealthCare","place":["Budapest, Hungary"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongxu","family":"Yang","sequence":"additional","affiliation":[{"name":"Science and Technology Organization, GE HealthCare","place":["Eindhoven, Netherlands"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,23]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"2293","DOI":"10.1021\/cr980440x","article-title":"Gadolinium(III) chelates as MRI contrast agents: structure, dynamics, and applications","volume":"99","author":"Caravan","year":"1999","journal-title":"Chem Rev"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1002\/wnan.1198","article-title":"Gadolinium-based contrast agents for magnetic resonance cancer imaging","volume":"5","author":"Zhou","year":"2013","journal-title":"Wiley Interdiscip Rev Nanomed Nanobiotechnol"},{"key":"B3","doi-asserted-by":"publisher","first-page":"407","DOI":"10.1002\/jmri.10176","article-title":"Dynamic contrast-enhanced MRI in clinical oncology: current status and future directions","volume":"16","author":"Padhani","year":"2002","journal-title":"J Magn Reson Imaging"},{"key":"B4","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1593\/tlo.13838","article-title":"Variations of dynamic contrast-enhanced magnetic resonance imaging in evaluation of breast cancer therapy response: a multicenter data analysis challenge","volume":"7","author":"Huang","year":"2014","journal-title":"Transl Oncol"},{"key":"B5","doi-asserted-by":"publisher","first-page":"63","DOI":"10.1016\/S1470-2045(06)71012-9","article-title":"Dynamic contrast-enhanced MRI as a predictor of tumour response to radiotherapy","volume":"8","author":"Zahra","year":"2007","journal-title":"Lancet Oncol"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1059","DOI":"10.1136\/amiajnl-2012-001460","article-title":"Dynamic contrast-enhanced MRI-based biomarkers of therapeutic response in triple-negative breast cancer","volume":"20","author":"Golden","year":"2013","journal-title":"J Am Med Inform Assoc"},{"key":"B7","doi-asserted-by":"publisher","first-page":"1288","DOI":"10.1007\/s11547-020-01220-z","article-title":"Dynamic contrast-enhanced MRI in oncology: how we do it","volume":"125","author":"Petralia","year":"2020","journal-title":"Radiol Med"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1149","DOI":"10.1093\/brain\/120.7.1149","article-title":"Sensitivity of contrast enhanced MRI in multiple sclerosis. Effects of gadolinium dose, magnetization transfer contrast and delayed imaging","volume":"120","author":"Silver","year":"1997","journal-title":"Brain"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1493","DOI":"10.1212\/WNL.0b013e3182553bd2","article-title":"Characterizing contrast-enhancing and re-enhancing lesions in multiple sclerosis","volume":"78","author":"Campbell","year":"2012","journal-title":"Neurology"},{"key":"B10","doi-asserted-by":"publisher","first-page":"591","DOI":"10.1007\/s00234-005-1383-7","article-title":"Contrast-enhanced FLAIR in the early diagnosis of infectious meningitis","volume":"47","author":"Splendiani","year":"2005","journal-title":"Neuroradiology"},{"key":"B11","doi-asserted-by":"publisher","first-page":"324","DOI":"10.1602\/neurorx.2.2.324","article-title":"Neuroimaging of infections","volume":"2","author":"Kastrup","year":"2005","journal-title":"NeuroRx"},{"key":"B12","doi-asserted-by":"publisher","first-page":"26","DOI":"10.4103\/2394-8108.164996","article-title":"Recent advances in magnetic resonance imaging for stroke diagnosis","volume":"1","author":"Rastogi","year":"2015","journal-title":"Brain Circ"},{"key":"B13","doi-asserted-by":"publisher","first-page":"3840","DOI":"10.1007\/s00330-018-5323-4","article-title":"Three-dimensional black-blood contrast-enhanced MRI improves detection of intraluminal thrombi in patients with acute ischaemic stroke","volume":"28","author":"Jang","year":"2018","journal-title":"Eur Radiol"},{"key":"B14","doi-asserted-by":"publisher","first-page":"2873","DOI":"10.1161\/01.CIR.0000044389.51236.91","article-title":"Visualization of ventricular thrombi with contrast-enhanced magnetic resonance imaging in patients with ischemic heart disease","volume":"106","author":"Mollet","year":"2002","journal-title":"Circulation"},{"key":"B15","doi-asserted-by":"publisher","first-page":"653","DOI":"10.1148\/radiol.2323031558","article-title":"Contrast-enhanced MR imaging of the heart: overview of the literature","volume":"232","author":"Edelman","year":"2004","journal-title":"Radiology"},{"key":"B16","doi-asserted-by":"publisher","first-page":"687","DOI":"10.1148\/radiology.184.3.1324509","article-title":"Focal liver disease: comparison of dynamic contrast-enhanced CT and T2-weighted fat-suppressed, FLASH, and dynamic gadolinium-enhanced MR imaging at 1.5T","volume":"184","author":"Semelka","year":"1992","journal-title":"Radiology"},{"key":"B17","doi-asserted-by":"publisher","first-page":"559","DOI":"10.1007\/s00330-004-2236-1","article-title":"Hepatobiliary contrast agents for contrast-enhanced MRI of the liver: properties, clinical development and applications","volume":"14","author":"Reimer","year":"2004","journal-title":"Eur Radiol"},{"key":"B18","doi-asserted-by":"publisher","first-page":"351","DOI":"10.2214\/AJR.04.1798","article-title":"Psoriatic arthritis and rheumatoid arthritis: findings in contrast-enhanced MRI","volume":"187","author":"Schoellnast","year":"2006","journal-title":"Am J Roentgenol"},{"key":"B19","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1148\/radiol.2431062144","article-title":"Nephrogenic systemic fibrosis: risk factors and incidence estimation","volume":"243","author":"Sadowski","year":"2007","journal-title":"Radiology"},{"key":"B20","doi-asserted-by":"publisher","first-page":"260","DOI":"10.1038\/sj.ki.5002338","article-title":"Gadolinium and nephrogenic systemic fibrosis","volume":"72","author":"Grobner","year":"2007","journal-title":"Kidney Int"},{"key":"B21","doi-asserted-by":"publisher","first-page":"517","DOI":"10.1148\/radiol.2018181151","article-title":"Gadolinium retention: a research roadmap from the 2018 NIH\/ACR\/RSNA workshop on gadolinium chelates","volume":"289","author":"McDonald","year":"2018","journal-title":"Radiology"},{"key":"B22","doi-asserted-by":"publisher","first-page":"1853","DOI":"10.1001\/jama.2018.13362","article-title":"Gadolinium retention after contrast-enhanced MRI","volume":"320","author":"Levine","year":"2018","journal-title":"J Am Med Assoc"},{"key":"B23","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1097\/RLI.0000000000000404","article-title":"Evaluation of gadolinium retention after serial administrations of a macrocyclic gadolinium-based contrast agent (Gadobutrol): a single-institution experience with 189 patients","volume":"53","author":"Yoo","year":"2018","journal-title":"Invest Radiol"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.aca.2012.12.007","article-title":"Determination of gadolinium-based MRI contrast agents in biological and environmental samples: a review","volume":"764","author":"Telgmann","year":"2013","journal-title":"Anal Chim Acta"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1523","DOI":"10.1002\/etc.4116","article-title":"Gadolinium as a new emerging contaminant of aquatic environments","volume":"37","author":"Rogowska","year":"2018","journal-title":"Environ Toxicol Chem"},{"key":"B26","author":"Lebel","year":""},{"key":"B27","doi-asserted-by":"publisher","first-page":"101069","DOI":"10.1016\/j.jocmr.2024.101069","article-title":"Clinical utility of a rapid two-dimensional balanced steady-state free precession sequence with deep learning reconstruction","volume":"26","author":"Eyre","year":"2024","journal-title":"J Cardiovasc Magn Reson"},{"key":"B28","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1097\/RMR.0000000000000237","article-title":"Deep learning AI applications in the imaging of glioma","volume":"29","author":"Zlochower","year":"2020","journal-title":"Top Magn Reson Imaging"},{"key":"B29","first-page":"91","article-title":"Efficient and accurate MRI super-resolution using a generative adversarial network and 3D multi-level densely connected network","author":"Chen","year":"2018"},{"key":"B30","first-page":"1","article-title":"Zero-shot bias correction: efficient MR image inhomogeneity reduction without any data","author":"Yang","year":"2025"},{"key":"B31","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1109\/JBHI.2021.3097721","article-title":"Task-induced pyramid and attention GAN for multimodal brain image imputation and classification in Alzheimer\u2019s disease","volume":"26","author":"Gao","year":"2022","journal-title":"IEEE J Biomed Health Inform"},{"key":"B32","doi-asserted-by":"publisher","first-page":"330","DOI":"10.1002\/jmri.25970","article-title":"Deep learning enables reduced gadolinium dose for contrast-enhanced brain MRI","volume":"48","author":"Gong","year":"2018","journal-title":"J Magn Reson Imaging"},{"key":"B33","doi-asserted-by":"publisher","first-page":"1687","DOI":"10.1002\/mrm.28808","article-title":"A generic deep learning model for reduced gadolinium dose in contrast-enhanced brain MRI","volume":"86","author":"Pasumarthi","year":"2021","journal-title":"Magn Reson Med"},{"key":"B34","doi-asserted-by":"publisher","first-page":"746","DOI":"10.1097\/RLI.0000000000000983","article-title":"Artificial intelligence to reduce or eliminate the need for gadolinium-based contrast agents in brain and cardiac MRI: a literature review","volume":"58","author":"Mallio","year":"2023","journal-title":"Invest Radiol"},{"key":"B35","doi-asserted-by":"publisher","first-page":"848","DOI":"10.1002\/jmri.29095","article-title":"Reducing gadolinium contrast with artificial intelligence","volume":"60","author":"Tsui","year":"2024","journal-title":"J Magn Reson Imaging"},{"key":"B36","doi-asserted-by":"publisher","first-page":"1761","DOI":"10.1007\/s00330-024-11047-1","article-title":"Generating synthetic high-resolution spinal STIR and T1w images from T2w FSE and low-resolution axial Dixon","volume":"35","author":"Graf","year":"2025","journal-title":"Eur Radiol"},{"key":"B37","first-page":"1","article-title":"T1w MRI to T2w MRI image synthesis using ssim-cyclegan","author":"Meharban","year":"2024"},{"key":"B38","doi-asserted-by":"publisher","first-page":"401","DOI":"10.1109\/TBME.2022.3192309","article-title":"Synthesizing MR image contrast enhancement using 3D high-resolution ConvNets","volume":"70","author":"Chen","year":"2023","journal-title":"IEEE Trans Biomed Eng"},{"key":"B39","doi-asserted-by":"publisher","first-page":"653","DOI":"10.1097\/RLI.0000000000000583","article-title":"Can virtual contrast enhancement in brain MRI replace gadolinium? A feasibility study","volume":"54","author":"Kleesiek","year":"2019","journal-title":"Invest Radiol"},{"key":"B40","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1016\/j.radonc.2022.05.029","article-title":"Dedicated isotropic 3-D T1 SPACE sequence imaging for radiosurgery planning improves brain metastases detection and reduces the risk of intracranial relapse","volume":"173","author":"Kutuk","year":"2022","journal-title":"Radiother Oncol"},{"key":"B41","doi-asserted-by":"publisher","first-page":"361","DOI":"10.1007\/s00234-012-1127-4","article-title":"Evaluation of dynamic contrast-enhanced T1-weighted perfusion MRI in the differentiation of tumor recurrence from radiation necrosis","volume":"55","author":"Larsen","year":"2013","journal-title":"Neuroradiology"},{"key":"B42","doi-asserted-by":"publisher","first-page":"40","DOI":"10.21037\/cco.2017.06.28","article-title":"Advanced magnetic resonance imaging in glioblastoma: a review","volume":"6","author":"Shukla","year":"2017","journal-title":"Chin Clin Oncol"},{"key":"B43","doi-asserted-by":"publisher","first-page":"170117","DOI":"10.1038\/sdata.2017.117","article-title":"Advancing the cancer genome atlas glioma MRI collections with expert segmentation labels and radiomic features","volume":"4","author":"Bakas","year":"2017","journal-title":"Sci Data"},{"key":"B44","doi-asserted-by":"publisher","first-page":"1993","DOI":"10.1109\/TMI.2014.2377694","article-title":"The multimodal brain tumor image segmentation benchmark (BRATS)","volume":"34","author":"Menze","year":"2015","journal-title":"IEEE Trans Med Imaging"},{"key":"B45","author":"Bakas","year":""},{"key":"B46","author":"Bakas","year":""},{"key":"B47","article-title":"Pytorch: an imperative style, high-performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"B48","author":"Cardoso","year":""},{"key":"B49","doi-asserted-by":"publisher","first-page":"657","DOI":"10.59275\/j.melba.2024-d61g","article-title":"Cones: conditional neural fields with shift modulation for multi-sequence MRI translation","volume":"2","author":"Chen","year":"2023","journal-title":"J Mach Learn Biomed Imaging"},{"key":"B50","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume":"33","author":"Ho","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"B51","first-page":"623","article-title":"Towards performant and reliable undersampled MR reconstruction via diffusion model sampling","author":"Peng","year":"2022"},{"key":"B52","first-page":"10021","article-title":"LION: latent point diffusion models for 3D shape generation","volume-title":"Advances in Neural Information Processing Systems (NeurIPS)","author":"Zeng","year":"2022"},{"key":"B53","author":"Lyu","year":""},{"key":"B54","author":"Chen","year":""},{"key":"B55","first-page":"311","article-title":"3D MRI brain tumor segmentation using autoencoder regularization","author":"Myronenko","year":"2018"},{"key":"B56","author":"Loshchilov","year":""},{"key":"B57","doi-asserted-by":"publisher","first-page":"e70062","DOI":"10.1002\/nbm.70062","article-title":"Exploring the limitations of virtual contrast prediction in brain tumor imaging: a study of generalization across tumor types and patient populations","volume":"38","author":"Caragliano","year":"2025","journal-title":"NMR Biomed"}],"container-title":["Frontiers in Radiology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2026.1931738\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T05:49:35Z","timestamp":1790142575000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2026.1931738\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,23]]},"references-count":57,"alternative-id":["10.3389\/fradi.2026.1931738"],"URL":"https:\/\/doi.org\/10.3389\/fradi.2026.1931738","relation":{},"ISSN":["2673-8740"],"issn-type":[{"value":"2673-8740","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,23]]},"article-number":"1931738"}}