{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T05:21:11Z","timestamp":1788153671505,"version":"build-2803163510"},"reference-count":57,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T00:00:00Z","timestamp":1783555200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T00:00:00Z","timestamp":1783555200000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["CCF\u20102239687"],"award-info":[{"award-number":["CCF\u20102239687"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["2019844"],"award-info":[{"award-number":["2019844"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Magnetic Resonance in Med"],"published-print":{"date-parts":[[2026,11]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To develop a data\u2010efficient strategy for accelerated MRI reconstruction with Diffusion Probabilistic Generative Models (DPMs) that enables faster scan times in clinical stroke MRI when only limited fully\u2010sampled data are available.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Our simple training strategy first pre\u2010trains a DPM on a large, diverse collection of publicly available fastMRI brain data and then fine\u2010tunes on a small target dataset using carefully selected learning rates and fine\u2010tuning durations. The approach is evaluated on controlled fastMRI experiments and on clinical stroke MRI data with a blinded clinical reader study.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>DPMs pre\u2010trained on 4000 non\u2010FLAIR subjects and fine\u2010tuned on FLAIR data from only 20 target subjects achieve reconstruction performance comparable to models trained with substantially more target\u2010domain FLAIR data across multiple acceleration factors. Moderate fine\u2010tuning with a reduced learning rate yields improved performance, while insufficient or excessive fine\u2010tuning degrades reconstruction quality. In a blinded reader study of 80 subjects at a single clinical site, images reconstructed from  accelerated data using the proposed approach are rated comparably to standard\u2010of\u2010care on the image quality and structural delineation metrics defined in this work.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Large\u2010scale pre\u2010training combined with targeted fine\u2010tuning can enable DPM\u2010based MRI reconstruction for our data\u2010constrained, accelerated clinical stroke MRI application. In the single\u2010site settings evaluated here, the proposed approach reduces the need for large application\u2010specific datasets while maintaining clinically acceptable image quality, providing preliminary evidence for pre\u2010trained and fine\u2010tuned diffusion models as a strategy for accelerated MRI in targeted applications.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mrm.70501","type":"journal-article","created":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T16:08:08Z","timestamp":1783613288000},"page":"2396-2411","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Accelerating Stroke\n                    <scp>MRI<\/scp>\n                    With Diffusion Probabilistic Models Through Large\u2010Scale Pre\u2010Training and Target\u2010Specific Fine\u2010Tuning"],"prefix":"10.1002","volume":"96","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2194-1945","authenticated-orcid":false,"given":"Yamin","family":"Arefeen","sequence":"first","affiliation":[{"name":"Chandra Family Department of Electrical and Computer Engineering The University of Texas at Austin  Austin Texas USA"},{"name":"Department of Imaging Physics MD Anderson Cancer Center  Houston Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sidharth","family":"Kumar","sequence":"additional","affiliation":[{"name":"Chandra Family Department of Electrical and Computer Engineering The University of Texas at Austin  Austin Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Steven","family":"Warach","sequence":"additional","affiliation":[{"name":"Dell Medical School Department of Neurology The University of Texas at Austin  Austin Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hamidreza","family":"Saber","sequence":"additional","affiliation":[{"name":"Dell Medical School Department of Neurology The University of Texas at Austin  Austin Texas USA"},{"name":"Dell Medical School Department of Neurosurgery The University of Texas at Austin  Austin Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonathan Isaac","family":"Tamir","sequence":"additional","affiliation":[{"name":"Chandra Family Department of Electrical and Computer Engineering The University of Texas at Austin  Austin Texas USA"},{"name":"Dell Medical School Department of Diagnostic Medicine The University of Texas at Austin  Austin Texas USA"},{"name":"Oden Institute for Computational Engineering and Sciences The University of Texas at Austin  Austin Texas USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,7,9]]},"reference":[{"key":"e_1_2_8_2_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10334-024-01173-8"},{"key":"e_1_2_8_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2022.3215288"},{"key":"e_1_2_8_4_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2021.3075856"},{"key":"e_1_2_8_5_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2025.103485"},{"key":"e_1_2_8_6_1","unstructured":"A. D.Desai A. M.Schmidt E. B.Rubin et al. \u201cSkm\u2010Tea: A Dataset for Accelerated MRI Reconstruction With Dense Image Labels for Quantitative Clinical Evaluation \u201dAdvances in Neural Information Processing Systems (Datasets and Benchmarks Track)(2021)."},{"key":"e_1_2_8_7_1","unstructured":"R.Bommasani D. A.Hudson E.Adeli et al. \u201cOn the Opportunities and Risks of Foundation Models \u201darXiv(2021)."},{"key":"e_1_2_8_8_1","doi-asserted-by":"publisher","DOI":"10.1007\/s13042-024-02443-6"},{"key":"e_1_2_8_9_1","unstructured":"W. X.Zhao K.Zhou J.Li et al. \u201cA Survey of Large Language Models \u201darXiv(2023)."},{"key":"e_1_2_8_10_1","unstructured":"A.Radford J. W.Kim C.Hallacy et al. \u201cLearning Transferable Visual Models From Natural Language Supervision \u201dProceedings of Machine Learning Research(2021)."},{"key":"e_1_2_8_11_1","doi-asserted-by":"crossref","unstructured":"A.Kirillov E.Mintun N.Ravi et al. \u201cSegment Anything \u201dIEEE\/CVF International Conference on Computer Vision (ICCV)(2023).","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"e_1_2_8_12_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-024-02244-3"},{"key":"e_1_2_8_13_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-023-02504-3"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06555-x"},{"key":"e_1_2_8_15_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-05881-4"},{"key":"e_1_2_8_16_1","first-page":"67","article-title":"Imaging in Acute Stroke","volume":"12","author":"Birenbaum D.","year":"2011","journal-title":"Western Journal of Emergency Medicine"},{"key":"e_1_2_8_17_1","doi-asserted-by":"publisher","DOI":"10.1161\/STROKEAHA.123.045154"},{"key":"e_1_2_8_18_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.231938"},{"key":"e_1_2_8_19_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.27420"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.29036"},{"key":"e_1_2_8_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2023.3342156"},{"key":"e_1_2_8_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCI.2021.3097596"},{"key":"e_1_2_8_23_1","unstructured":"B.Yaman S. A. H.Hosseini andM.Ak\u00e7akaya \u201cZero\u2010Shot Self\u2010Supervised Learning for Mri Reconstruction \u201dInternational Conference on Learning Representations(2022)."},{"key":"e_1_2_8_24_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.28378"},{"key":"e_1_2_8_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCI.2023.3299212"},{"key":"e_1_2_8_26_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2025.3570226"},{"key":"e_1_2_8_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2025.103890"},{"key":"e_1_2_8_28_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.29759"},{"key":"e_1_2_8_29_1","unstructured":"A.Wang S.McDonagh andM.Davies \u201cBenchmarking Self\u2010Supervised Learning Methods for Accelerated MRI Reconstruction \u201darXiv(2025)."},{"key":"e_1_2_8_30_1","volume-title":"Test\u2010Time Training Can Close the Natural Distribution Shift Performance Gap in Deep Learning Based Compressed Sensing","author":"Darestani M. Z.","year":"2022"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102872"},{"key":"e_1_2_8_32_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2024.3524797"},{"key":"e_1_2_8_33_1","doi-asserted-by":"crossref","unstructured":"H.ChungandJ. C.Ye \u201cDeep Diffusion Image Prior for Efficient OOD Adaptation in 3D Inverse Problems \u201din European Conference on Computer Vision (ECCV)(2024).","DOI":"10.1007\/978-3-031-73226-3_25"},{"key":"e_1_2_8_34_1","doi-asserted-by":"crossref","unstructured":"Y.Korkmaz T.\u00c7ukur andV. M.Patel \u201cSelf\u2010Supervised MRI Reconstruction With Unrolled Diffusion Models \u201din Medical Image Computing and Computer Assisted Intervention (MICCAI)(2023).","DOI":"10.1007\/978-3-031-43999-5_47"},{"key":"e_1_2_8_35_1","unstructured":"J.Parmar S.Satheesh M.Patwary M.Shoeybi andB.Catanzaro \u201cReuse Don't Retrain: A Recipe for Continued Pretraining of Language Models \u201darXiv (2024)."},{"key":"e_1_2_8_36_1","unstructured":"K.LinandR.Heckel \u201cRobustness of Deep Learning for Accelerated Mri: Benefits of Diverse Training Data \u201dInternational Conference on Machine Learning(2024)."},{"key":"e_1_2_8_37_1","unstructured":"Y.Song J.Sohl\u2010Dickstein D. P.Kingma A.Kumar S.Ermon andB.Poole \u201cScore\u2010Based Generative Modeling Through Stochastic Differential Equations \u201dInternational Conference on Learning Representations(2021)."},{"key":"e_1_2_8_38_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102479"},{"key":"e_1_2_8_39_1","unstructured":"H.Chung J.Kim M. T.Mccann M. L.Klasky andJ. C.Ye \u201cDiffusion Posterior Sampling for General Noisy Inverse Problems \u201dInternational Conference on Learning Representations(2023)."},{"key":"e_1_2_8_40_1","unstructured":"A.Jalal M.Arvinte G.Daras E.Price A. G.Dimakis andJ. I.Tamir \u201cRobust Compressed Sensing Mri With Deep Generative Priors \u201dAdvances in Neural Information Processing Systems(2021)."},{"key":"e_1_2_8_41_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.29624"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/3554729"},{"key":"e_1_2_8_43_1","unstructured":"J.Ho A.Jain andP.Abbeel \u201cDenoising Diffusion Probabilistic Models \u201dAdvances in Neural Information Processing Systems(2020)."},{"key":"e_1_2_8_44_1","unstructured":"H.Zheng W.Chu B.Zhang et al. \u201cInversebench: Benchmarking Plug\u2010And\u2010Play Diffusion Priors for Inverse Problems in Physical Sciences \u201dInternational Conference on Learning Representations(2025)."},{"key":"e_1_2_8_45_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.28274"},{"key":"e_1_2_8_46_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2018.2887072"},{"key":"e_1_2_8_47_1","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1926"},{"key":"e_1_2_8_48_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.30585"},{"key":"e_1_2_8_49_1","volume-title":"Reconstruct Anything Model a Lightweight Foundation Model for Computational Imaging","author":"Terris M.","year":"2026"},{"key":"e_1_2_8_50_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.70374"},{"key":"e_1_2_8_51_1","doi-asserted-by":"publisher","DOI":"10.1002\/mrm.70072"},{"key":"e_1_2_8_52_1","unstructured":"M.Uecker F.Ong J. I.Tamir et al. \u201cBerkeley Advanced Reconstruction Toolbox \u201dProceedings of the International Society of Magnetic Resonance in Medicine(2015)."},{"key":"e_1_2_8_53_1","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.1611835114"},{"key":"e_1_2_8_54_1","unstructured":"E. J.Hu Y.Shen P.Wallis et al. \u201cLoRA: Low\u2010Rank Adaptation of Large Language Models \u201din International Conference on Learning Representations (ICLR)(2022)."},{"key":"e_1_2_8_55_1","doi-asserted-by":"crossref","unstructured":"R.Gandikota J.Materzy\u0144ska T.Zhou A.Torralba andD.Bau \u201cConcept Sliders: LoRA Adaptors for Precise Control in Diffusion Models \u201din European Conference on Computer Vision (ECCV)(2024).","DOI":"10.1007\/978-3-031-73661-2_10"},{"key":"e_1_2_8_56_1","unstructured":"N.Houlsby A.Giurgiu S.Jastrzebski et al. \u201cParameter\u2010Efficient Transfer Learning for NLP \u201din International Conference on Machine Learning (ICML)(2019)."},{"key":"e_1_2_8_57_1","unstructured":"D.Ha A. M.Dai andQ. V.Le \u201cHyperNetworks \u201din International Conference on Learning Representations (ICLR)(2017)."},{"key":"e_1_2_8_58_1","unstructured":"J.Yosinski J.Clune Y.Bengio andH.Lipson \u201cHow Transferable Are Features in Deep Neural Networks?\u201din Advances in Neural Information Processing Systems (NeurIPS)(2014)."}],"container-title":["Magnetic Resonance in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mrm.70501","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mrm.70501","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mrm.70501","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T04:54:00Z","timestamp":1788152040000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/mrm.70501"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,9]]},"references-count":57,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,11]]}},"alternative-id":["10.1002\/mrm.70501"],"URL":"https:\/\/doi.org\/10.1002\/mrm.70501","archive":["Portico"],"relation":{},"ISSN":["0740-3194","1522-2594"],"issn-type":[{"value":"0740-3194","type":"print"},{"value":"1522-2594","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,9]]},"assertion":[{"value":"2026-03-11","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-06-25","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-09","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}