{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,17]],"date-time":"2026-08-17T15:19:47Z","timestamp":1786979987366,"version":"build-2736575974"},"reference-count":28,"publisher":"Springer Science and Business Media LLC","issue":"2","license":[{"start":{"date-parts":[[2024,8,1]],"date-time":"2024-08-01T00:00:00Z","timestamp":1722470400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,8,1]],"date-time":"2024-08-01T00:00:00Z","timestamp":1722470400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur Radiol"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>To investigate the use of the score-based diffusion model to accelerate breast MRI reconstruction.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>We trained a score-based model on 9549 MRI examinations of the female breast and employed it to reconstruct undersampled MRI images with undersampling factors of 2, 5, and 20. Images were evaluated by two experienced radiologists who rated the images based on their overall quality and diagnostic value on an independent test set of 100 additional MRI examinations.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The score-based model produces MRI images of high quality and diagnostic value. Both T1- and T2-weighted MRI images could be reconstructed to a high degree of accuracy. Two radiologists rated the images as almost indistinguishable from the original images (rating 4 or 5 on a scale of 5) in 100% (radiologist 1) and 99% (radiologist 2) of cases when the acceleration factor was 2. This fraction dropped to 88% and 70% for an acceleration factor of 5 and to 5% and 21% with an extreme acceleration factor of 20.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Score-based models can reconstruct MRI images at high fidelity, even at comparatively high acceleration factors, but further work on a larger scale of images is needed to ensure that diagnostic quality holds.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Clinical relevance statement<\/jats:title>\n                    <jats:p>The number of MRI examinations of the breast is expected to rise with MRI screening recommended for women with dense breasts. Accelerated image acquisition methods can help in making this examination more accessible.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Key Points<\/jats:title>\n                    <jats:p>\n                      <jats:list list-type=\"bullet\">\n                        <jats:list-item>\n                          <jats:p>\n                            <jats:italic>Accelerating breast MRI reconstruction remains a significant challenge in clinical settings<\/jats:italic>\n                            .\n                          <\/jats:p>\n                        <\/jats:list-item>\n                        <jats:list-item>\n                          <jats:p>\n                            <jats:italic>Score-based diffusion models can achieve near-perfect reconstruction for moderate undersampling factors<\/jats:italic>\n                            .\n                          <\/jats:p>\n                        <\/jats:list-item>\n                        <jats:list-item>\n                          <jats:p>\n                            <jats:italic>Faster breast MRI scans with maintained image quality could revolutionize clinic workflows and patient experience<\/jats:italic>\n                            .\n                          <\/jats:p>\n                        <\/jats:list-item>\n                      <\/jats:list>\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00330-024-10853-x","type":"journal-article","created":{"date-parts":[[2024,8,1]],"date-time":"2024-08-01T11:02:38Z","timestamp":1722510158000},"page":"1092-1100","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["Accelerating breast MRI acquisition with generative AI models"],"prefix":"10.1007","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8684-1358","authenticated-orcid":false,"given":"Augustine","family":"Okolie","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Timm","family":"Dirrichs","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Luisa Charlotte","family":"Huck","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sven","family":"Nebelung","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Soroosh Tayebi","family":"Arasteh","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Teresa","family":"Nolte","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tianyu","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christiane Katharina","family":"Kuhl","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Truhn","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2024,8,1]]},"reference":[{"key":"10853_CR1","doi-asserted-by":"publisher","first-page":"377","DOI":"10.1002\/jmri.26654","volume":"50","author":"RM Mann","year":"2019","unstructured":"Mann RM, Kuhl CK, Moy L (2019) Contrast-enhanced MRI for breast cancer screening. J Magn Reson Imaging 50:377\u2013390","journal-title":"J Magn Reson Imaging"},{"key":"10853_CR2","doi-asserted-by":"publisher","first-page":"4036","DOI":"10.1007\/s00330-022-08617-6","volume":"32","author":"RM Mann","year":"2022","unstructured":"Mann RM, Athanasiou A, Baltzer PAT et al (2022) Breast cancer screening in women with extremely dense breasts: recommendations of the European Society of Breast Imaging (EUSOBI). Eur Radiol 32:4036\u20134045","journal-title":"Eur Radiol"},{"key":"10853_CR3","doi-asserted-by":"publisher","first-page":"952","DOI":"10.1002\/(SICI)1522-2594(199911)42:5<952::AID-MRM16>3.0.CO;2-S","volume":"42","author":"KP Pruessmann","year":"1999","unstructured":"Pruessmann KP, Weiger M, Scheidegger MB, Boesiger P (1999) Sense: sensitivity encoding for fast MRI. Magn Reson Med 42:952\u2013962","journal-title":"Magn Reson Med"},{"key":"10853_CR4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s42490-019-0006-z","volume":"1","author":"JC Ye","year":"2019","unstructured":"Ye JC (2019) Compressed sensing MRI: a review from signal processing perspective. BMC Biomed Eng 1:1\u201317","journal-title":"BMC Biomed Eng"},{"key":"10853_CR5","doi-asserted-by":"publisher","DOI":"10.1016\/j.cosrev.2020.100285","volume":"38","author":"GM Harshvardhan","year":"2020","unstructured":"Harshvardhan GM, Gourisaria MK, Pandey M, Rautaray SS (2020) A comprehensive survey and analysis of generative models in machine learning. Comput Sci Rev 38:100285","journal-title":"Comput Sci Rev"},{"key":"10853_CR6","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.abb7973","volume":"6","author":"T Han","year":"2020","unstructured":"Han T, Nebelung S, Haarburger C et al (2020) Breaking medical data sharing boundaries by using synthesized radiographs. Sci Adv 6:eabb7973","journal-title":"Sci Adv"},{"key":"10853_CR7","doi-asserted-by":"crossref","unstructured":"M\u00fcller-Franzes G, Huck L, Tayebi Arasteh S et al (2023) Using machine learning to reduce the need for contrast agents in breast MRI through synthetic images. Radiology 307:e222211","DOI":"10.1148\/radiol.222211"},{"key":"10853_CR8","unstructured":"Cui ZX, Cao C, Liu S et al (2022) Self-score: Self-supervised learning on score-based models for MRI reconstruction. arXiv preprint arXiv:2209.00835"},{"key":"10853_CR9","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102479","volume":"80","author":"H Chung","year":"2022","unstructured":"Chung H, Ye JC (2022) Score-based diffusion models for accelerated MRI. Med Image Anal 80:102479","journal-title":"Med Image Anal"},{"key":"10853_CR10","unstructured":"Song Y, Sohl-Dickstein J, Kingma DP, Kumar A, Ermon S, Poole B (2021) Score-based generative modeling through stochastic differential equations. In International Conference on Learning Representations. Retrieved from https:\/\/openreview.net\/forum?id=PxTIG12RRHS"},{"key":"10853_CR11","first-page":"12438","volume":"33","author":"Y Song","year":"2020","unstructured":"Song Y, Ermon S (2020) Improved techniques for training score-based generative models. Adv Neural Inf Process Syst 33:12438\u201312448","journal-title":"Adv Neural Inf Process Syst"},{"key":"10853_CR12","doi-asserted-by":"publisher","unstructured":"Song Y, Ermon S (2019) Generative modeling by estimating gradients of the data distribution. Adv Neural Inf Process Syst 32. https:\/\/doi.org\/10.48550\/arXiv.1907.05600","DOI":"10.48550\/arXiv.1907.05600"},{"key":"10853_CR13","first-page":"14938","volume":"34","author":"A Jalal","year":"2021","unstructured":"Jalal A, Arvinte M, Daras G et al (2021) Robust compressed sensing MRI with deep generative priors. Adv Neural Inf Process Syst 34:14938\u201314954","journal-title":"Adv Neural Inf Process Syst"},{"key":"10853_CR14","doi-asserted-by":"crossref","unstructured":"Cole EK, Ong F, Vasanawala SS, Pauly JM (2021). Fast unsupervised mri reconstruction without fully-sampled ground truth data using generative adversarial networks. In Proceedings of the IEEE\/CVF International Conference on Computer Vision (3988\u20133997)","DOI":"10.1109\/ICCVW54120.2021.00444"},{"key":"10853_CR15","unstructured":"Cole EK, Pauly JM, Vasanawala SS, Ong F (2020) Unsupervised MRI reconstruction with generative adversarial networks. arXiv preprint arXiv:2008.13065"},{"key":"10853_CR16","doi-asserted-by":"publisher","unstructured":"Song Y, Shu R, Kushman N, Ermon S (2018) Constructing unrestricted adversarial examples with generative models. Adv Neural Inf Process Syst 31. https:\/\/doi.org\/10.48550\/arXiv.1805.07894","DOI":"10.48550\/arXiv.1805.07894"},{"key":"10853_CR17","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-39278-0","volume":"13","author":"G M\u00fcller-Franzes","year":"2023","unstructured":"M\u00fcller-Franzes G, Niehues JM, Khader F et al (2023) A multimodal comparison of latent denoising diffusion probabilistic models and generative adversarial networks for medical image synthesis. Sci Rep 13:12098","journal-title":"Sci Rep"},{"key":"10853_CR18","doi-asserted-by":"publisher","first-page":"361","DOI":"10.1148\/radiol.2016161444","volume":"283","author":"CK Kuhl","year":"2017","unstructured":"Kuhl CK, Strobel K, Bieling H et al (2017) Supplemental breast MR imaging screening of women with average risk of breast cancer. Radiology 283:361\u2013370","journal-title":"Radiology"},{"key":"10853_CR19","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. Medical Image Computing and Computer-Assisted Intervention\u2013MICCAI 2015: 18th International Conference, Munich, Germany, October 5\u20139, 2015, Proceedings, Part III 18, 234\u2013241. Springer","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"10853_CR20","unstructured":"Hyv\u00e4rinen A, Dayan P (2005) Estimation of non-normalized statistical models by score matching. J Mach Learn Res 6:695\u2013709"},{"key":"10853_CR21","doi-asserted-by":"publisher","first-page":"1661","DOI":"10.1162\/NECO_a_00142","volume":"23","author":"P Vincent","year":"2011","unstructured":"Vincent P (2011) A connection between score matching and denoising autoencoders. Neural Comput 23:1661\u20131674","journal-title":"Neural Comput"},{"key":"10853_CR22","doi-asserted-by":"crossref","unstructured":"Hore A, Ziou D (2010) Image quality metrics: PSNR vs. SSIM. In 2010 20th international conference on pattern recognition (pp. 2366\u20132369). IEEE","DOI":"10.1109\/ICPR.2010.579"},{"key":"10853_CR23","doi-asserted-by":"publisher","first-page":"4275","DOI":"10.1007\/s00521-019-04332-4","volume":"32","author":"FDS Ribeiro","year":"2020","unstructured":"Ribeiro FDS, Caliva F, Swainson M et al (2020) Deep Bayesian self-training. Neural Comput Appl 32:4275\u20134291","journal-title":"Neural Comput Appl"},{"key":"10853_CR24","doi-asserted-by":"publisher","first-page":"1064","DOI":"10.1109\/TMI.2019.2930338","volume":"39","author":"A Mason","year":"2019","unstructured":"Mason A, Rioux J, Clarke SE et al (2019) Comparison of objective image quality metrics to expert radiologists\u2019 scoring of diagnostic quality of MR images. IEEE Trans Med Imaging 39:1064\u20131072","journal-title":"IEEE Trans Med Imaging"},{"key":"10853_CR25","doi-asserted-by":"publisher","first-page":"1202","DOI":"10.1002\/mrm.10171","volume":"47","author":"MA Griswold","year":"2002","unstructured":"Griswold MA, Jakob PM, Heidemann RM et al (2002) Generalized autocalibrating partially parallel acquisitions (GRAPPA). Magn Reson Med 47:1202\u20131210","journal-title":"Magn Reson Med"},{"key":"10853_CR26","doi-asserted-by":"crossref","unstructured":"Johnson PM, Lin DJ, Zbontar J et al (2023) Deep learning reconstruction enables prospectively accelerated clinical knee MRI. Radiology 307:e220425","DOI":"10.1148\/radiol.220425"},{"key":"10853_CR27","doi-asserted-by":"publisher","first-page":"373","DOI":"10.1148\/radiol.204097","volume":"303","author":"T Ueda","year":"2022","unstructured":"Ueda T, Ohno Y, Yamamoto K et al (2022) Deep learning reconstruction of diffusion-weighted MRI improves image quality for prostatic imaging. Radiology 303:373\u2013381","journal-title":"Radiology"},{"key":"10853_CR28","doi-asserted-by":"publisher","first-page":"4036","DOI":"10.1007\/s00330-022-08617-6","volume":"32","author":"RM Mann","year":"2022","unstructured":"Mann RM, Athanasiou A, Baltzer PA et al (2022) Breast cancer screening in women with extremely dense breasts recommendations of the European Society of Breast Imaging (EUSOBI). Eur Radiol 32:4036\u20134045","journal-title":"Eur Radiol"}],"container-title":["European Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-024-10853-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00330-024-10853-x\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-024-10853-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,30]],"date-time":"2025-01-30T17:53:43Z","timestamp":1738259623000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00330-024-10853-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,8,1]]},"references-count":28,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,2]]}},"alternative-id":["10853"],"URL":"https:\/\/doi.org\/10.1007\/s00330-024-10853-x","relation":{},"ISSN":["1432-1084"],"issn-type":[{"value":"1432-1084","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,8,1]]},"assertion":[{"value":"25 August 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 April 2024","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 June 2024","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 August 2024","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Compliance with ethical standards"}},{"value":"The scientific guarantor of this publication is Dr. Med. Daniel Truhn.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Guarantor"}},{"value":"The authors of this manuscript declare no conflict of interest.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"Dr. Augustine Okolie, an author with an extensive background in statistical methodologies, acted as the statistical guarantor for this study.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Statistics and biometry"}},{"value":"Only if the study is on human subjects: Written informed consent was waived by the Institutional Review Board.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Informed consent"}},{"value":"Institutional Review Board approval was obtained.","order":6,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"We analyzed 9751 MRI exams from 5086 patients between 2010 and 2019. Conversely, a prior study published by M\u00fcller-Franzes et al [\n                      \n                      ] investigated using GANs to generate contrast-enhanced breast MRIs from non-enhanced or low-contrast images.\n                      Overlap:\n                      Both studies used the same cohort of 9751 breast MRI examinations from 5086 patients spanning from January 2010 to December 2019.\n                      Distinctive Features:\n                      Objective and Approach: Our study (\u201cAccelerating Breast MRI Acquisition with Generative AI Models\u201d) centers on the application of score-based diffusion models to accelerate breast MRI reconstruction.\u00a0In contrast the study by\u00a0M\u00fcller-Franzes et al [\n                      \n                      ] investigates the use of generative adversarial networks (GANs) to produce synthetic contrast-enhanced MRI images either from unenhanced photos or virtual low-contrast-enhanced images. Research Question: Our study questions how much the scan times can be reduced using score-based diffusion models without compromising the quality of breast MRI images. The previously published paper investigates whether GANs can recreate contrast-enhanced breast MRI scans from unenhanced and virtually low-contrast-enhanced images. Outcome Metrics: In our study, the focus is on ratings of reconstructed images at different undersampling factors. The previously published paper measured the ability of radiologists to distinguish between real and synthesized contrast-enhanced images and compared the appearance and conspicuity of enhancing lesions on real vs. synthesized images. Clinical Implications: Our study suggests the possibility of substantially reducing MRI scan times without compromising\u00a0image quality. The previously published paper emphasized the potential for breast MRI with a reduced contrast agent dose.","order":7,"name":"Ethics","group":{"name":"EthicsHeading","label":"Study subjects or cohorts overlap"}},{"order":8,"name":"Ethics","group":{"name":"EthicsHeading","label":"Methodology"}}]}}