{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,21]],"date-time":"2026-05-21T17:17:02Z","timestamp":1779383822730,"version":"3.53.1"},"reference-count":37,"publisher":"MIT Press","license":[{"start":{"date-parts":[[2025,11,7]],"date-time":"2025-11-07T00:00:00Z","timestamp":1762473600000},"content-version":"vor","delay-in-days":310,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001665","name":"French National Research Agency","doi-asserted-by":"crossref","award":["ANR-20-THIA-0018"],"award-info":[{"award-number":["ANR-20-THIA-0018"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001665","name":"French National Research Agency","doi-asserted-by":"crossref","award":["ANR-21-RHUS-0014"],"award-info":[{"award-number":["ANR-21-RHUS-0014"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001665","name":"French National Research Agency","doi-asserted-by":"crossref","award":["ANR-10-COHO-002 OFSEP"],"award-info":[{"award-number":["ANR-10-COHO-002 OFSEP"]}],"id":[{"id":"10.13039\/501100001665","id-type":"DOI","asserted-by":"crossref"}]},{"name":"NIH","award":["NIH R01 LM013608"],"award-info":[{"award-number":["NIH R01 LM013608"]}]}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2025,11,26]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Detecting multiple sclerosis lesions in spinal cord MRI is a critical but complex task for radiologists and neurologists. While deep learning models have shown promise for this task, training these models requires large manually annotated datasets, which are time consuming and expensive to create. To reduce the annotation burden, this study proposes LesionSCynth, a parametric framework for synthesising hyperintense lesions in spinal cord MRI. Based on analysis of the intensity distribution of real lesions in sagittal T2-weighted acquisitions, LesionSCynth generates realistic synthetic lesions that can augment small annotated datasets. Segmentation models trained on a combination of these synthetic lesions and 17 real acquisitions achieved notably better performance than those trained on the real lesions alone (0.52 vs. 0.46 FROC, respectively). Notably, models trained with LesionSCynth lesions achieved similar detection performance to a model trained on eight times more real lesions (0.55 FROC for both). Moreover, LesionSCynth outperformed two leading lesion synthesis methods, LesionMix (0.43) and CarveMix (0.41), in low-data regimes. These findings position LesionSCynth as a practical and effective solution for reducing the annotation burden while improving multiple sclerosis (MS) lesion detection and segmentation in spinal cord MRI.<\/jats:p>","DOI":"10.1162\/imag.a.1029","type":"journal-article","created":{"date-parts":[[2025,11,7]],"date-time":"2025-11-07T21:10:07Z","timestamp":1762549807000},"update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":1,"title":["LesionSCynth: A simple parametric lesion synthesis method to improve spinal cord lesion segmentation in low-data scenarios"],"prefix":"10.1162","volume":"3","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2363-5506","authenticated-orcid":false,"given":"Ricky","family":"Walsh","sequence":"first","affiliation":[{"name":"Univ Rennes, Inria, CNRS, Inserm, IRISA UMR 6074, Empenn, Rennes, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5416-0219","authenticated-orcid":false,"given":"Prabhjot","family":"Kaur","sequence":"additional","affiliation":[{"name":"Computational Radiology Laboratory, Boston Children\u2019s Hospital and Harvard Medical School, Boston, MA, United States"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5155-2644","authenticated-orcid":false,"given":"Davood","family":"Karimi","sequence":"additional","affiliation":[{"name":"Computational Radiology Laboratory, Boston Children\u2019s Hospital and Harvard Medical School, Boston, MA, United States"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anne","family":"Kerbrat","sequence":"additional","affiliation":[{"name":"Univ Rennes, Inria, CNRS, Inserm, IRISA UMR 6074, Empenn, Rennes, France"},{"name":"Department of Neurology, Rennes University Hospital, Rennes, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7659-3880","authenticated-orcid":false,"given":"Simon K.","family":"Warfield","sequence":"additional","affiliation":[{"name":"Computational Radiology Laboratory, Boston Children\u2019s Hospital and Harvard Medical School, Boston, MA, United States"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8788-2856","authenticated-orcid":false,"given":"Francesca","family":"Galassi","sequence":"additional","affiliation":[{"name":"Univ Rennes, Inria, CNRS, Inserm, IRISA UMR 6074, Empenn, Rennes, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4161-7093","authenticated-orcid":false,"given":"Benoit","family":"Comb\u00e8s","sequence":"additional","affiliation":[{"name":"Univ Rennes, Inria, CNRS, Inserm, IRISA UMR 6074, Empenn, Rennes, France"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"281","published-online":{"date-parts":[[2025,11,26]]},"reference":[{"key":"2025112614554962000_IMAG.a.1029-b1","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-58171-7_8","article-title":"LesionMix: A lesion-level data augmentation method for medical image segmentation","volume-title":"MICCAI","author":"Basaran","year":"2024"},{"key":"2025112614554962000_IMAG.a.1029-b2","article-title":"Automatic multi-contrast MRI segmentation of spinal cord lesions","volume-title":"10th Annual Americas Committee for Treatment and Research in Multiple Sclerosis (ACTRIMS) Forum","author":"Benveniste","year":"2025"},{"issue":"6","key":"2025112614554962000_IMAG.a.1029-b3","first-page":"e230514","article-title":"Addressing the generalizability of AI in radiology using a novel data augmentation framework with synthetic patient image data: Proof-of-concept and external validation for classification tasks in multiple sclerosis","volume":"6","author":"Brugnara","year":"2024","journal-title":"Radiology: Artificial Intelligence"},{"key":"2025112614554962000_IMAG.a.1029-b4","doi-asserted-by":"crossref","first-page":"77","DOI":"10.1016\/j.neuroimage.2016.12.064","article-title":"Longitudinal multiple sclerosis lesion segmentation: Resource and challenge","volume":"148","author":"Carass","year":"2017","journal-title":"NeuroImage"},{"key":"2025112614554962000_IMAG.a.1029-b5","article-title":"Towards generalizable tumor synthesis","author":"Chen","year":"2024","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition"},{"issue":"1","key":"2025112614554962000_IMAG.a.1029-b6","doi-asserted-by":"crossref","first-page":"219","DOI":"10.1038\/s41597-021-00941-8","article-title":"Open-access quantitative MRI data of the spinal cord and reproducibility across participants, sites and manufacturers","volume":"8","author":"Cohen-Adad","year":"2021","journal-title":"Scientific Data"},{"key":"2025112614554962000_IMAG.a.1029-b7","doi-asserted-by":"crossref","first-page":"740248","DOI":"10.3389\/fmed.2021.740248","article-title":"A clinically-compatible workflow for computer-aided assessment of brain disease activity in multiple sclerosis patients","volume":"8","author":"Comb\u00e8s","year":"2021","journal-title":"Frontiers in Medicine"},{"key":"2025112614554962000_IMAG.a.1029-b8","article-title":"MSSEG challenge proceedings: Multiple sclerosis lesions segmentation challenge using a data management and processing infrastructure","author":"Commowick","year":"2016","journal-title":"MICCAI"},{"key":"2025112614554962000_IMAG.a.1029-b9","article-title":"MSSEG-2 challenge proceedings: Multiple sclerosis new lesions segmentation challenge using a data management and processing infrastructure","author":"Commowick","year":"2021","journal-title":"MICCAI 2021\u201424th International Conference on Medical Image Computing and Computer Assisted Intervention"},{"issue":"Pt A","key":"2025112614554962000_IMAG.a.1029-b10","doi-asserted-by":"crossref","first-page":"24","DOI":"10.1016\/j.neuroimage.2016.10.009","article-title":"SCT: Spinal Cord Toolbox, an open-source software for processing spinal cord MRI data","volume":"145","author":"De Leener","year":"2017","journal-title":"NeuroImage"},{"key":"2025112614554962000_IMAG.a.1029-b11","doi-asserted-by":"crossref","first-page":"103579","DOI":"10.1016\/j.compbiomed.2019.103579","article-title":"Simulated perfusion MRI data to boost training of convolutional neural networks for lesion fate prediction in acute stroke","volume":"116","author":"Debs","year":"2020","journal-title":"Computers in Biology and Medicine"},{"key":"2025112614554962000_IMAG.a.1029-b12","doi-asserted-by":"crossref","unstructured":"Ferreira, A., Solak, N., Li, J., Dammann, P., Kleesiek, J., Alves, V., & Egger, J. (2024, July). How we won BraTS 2023 Adult Glioma challenge? Just faking it! Enhanced Synthetic Data Augmentation and Model Ensemble for brain tumour segmentation. arXiv. https:\/\/doi.org\/10.48550\/arXiv.2402.17317","DOI":"10.1007\/978-3-031-76163-8_8"},{"key":"2025112614554962000_IMAG.a.1029-b13","first-page":"2672","article-title":"Generative adversarial nets","volume":"27","author":"Goodfellow","year":"2014","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2025112614554962000_IMAG.a.1029-b14","doi-asserted-by":"crossref","first-page":"901","DOI":"10.1016\/j.neuroimage.2018.09.081","article-title":"Automatic segmentation of the spinal cord and intramedullary multiple sclerosis lesions with convolutional neural networks","volume":"184","author":"Gros","year":"2019","journal-title":"NeuroImage"},{"key":"2025112614554962000_IMAG.a.1029-b15","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume":"33","author":"Ho","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"2","key":"2025112614554962000_IMAG.a.1029-b16","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1038\/s41592-020-01008-z","article-title":"nnU-Net: A self-configuring method for deep learning-based biomedical image segmentation","volume":"18","author":"Isensee","year":"2021","journal-title":"Nature Methods"},{"key":"2025112614554962000_IMAG.a.1029-b17","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-72114-4_47","article-title":"nnU-Net revisited: A call for rigorous validation in 3D medical image segmentation","volume-title":"MICCAI","author":"Isensee","year":"2024"},{"issue":"10422","key":"2025112614554962000_IMAG.a.1029-b18","doi-asserted-by":"crossref","first-page":"183","DOI":"10.1016\/S0140-6736(23)01473-3","article-title":"Multiple sclerosis","volume":"403","author":"Jakimovski","year":"2024","journal-title":"The Lancet"},{"key":"2025112614554962000_IMAG.a.1029-b19","article-title":"Automatic segmentation of spinal cord lesions in MS: A robust tool for axial T2-weighted MRI scans","author":"Karthik","year":"2025","journal-title":"Imaging Neuroscience"},{"issue":"6","key":"2025112614554962000_IMAG.a.1029-b20","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1038\/nrneurol.2015.80","article-title":"Spinal cord MRI in multiple sclerosis\u2014Diagnostic, prognostic and clinical value","volume":"11","author":"Kearney","year":"2015","journal-title":"Nature Reviews Neurology"},{"issue":"1","key":"2025112614554962000_IMAG.a.1029-b21","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1007\/s12021-017-9348-7","article-title":"A novel public MR image dataset of multiple sclerosis patients with lesion segmentations based on multi-rater consensus","volume":"16","author":"Lesjak","year":"2018","journal-title":"Neuroinformatics"},{"issue":"10","key":"2025112614554962000_IMAG.a.1029-b22","doi-asserted-by":"crossref","first-page":"5954","DOI":"10.1007\/s00330-025-11541-0","article-title":"Evaluation of a deep learning segmentation tool to help detect spinal cord lesions from combined T2 and STIR acquisitions in people with multiple sclerosis","volume":"35","author":"Lod\u00e9","year":"2025","journal-title":"European Radiology"},{"issue":"2","key":"2025112614554962000_IMAG.a.1029-b23","doi-asserted-by":"crossref","first-page":"406","DOI":"10.1007\/s00415-019-09591-8","article-title":"3D PSIR MRI at 3 Tesla improves detection of spinal cord lesions in multiple sclerosis","volume":"267","author":"Mirafzal","year":"2020","journal-title":"Journal of Neurology"},{"key":"2025112614554962000_IMAG.a.1029-b24","doi-asserted-by":"crossref","first-page":"106236","DOI":"10.1016\/j.cmpb.2021.106236","article-title":"TorchIO: A Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning","volume":"208","author":"P\u00e9rez-Garc\u00eda","year":"2021","journal-title":"Computer Methods and Programs in Biomedicine"},{"issue":"3","key":"2025112614554962000_IMAG.a.1029-b25","doi-asserted-by":"crossref","first-page":"780","DOI":"10.1016\/j.ejrad.2010.09.031","article-title":"Comparison of MRI sequences for evaluation of multiple sclerosis of the cervical spinal cord at 3T","volume":"80","author":"Philpott","year":"2011","journal-title":"European Journal of Radiology"},{"key":"2025112614554962000_IMAG.a.1029-b26","unstructured":"Rolnick, D., Veit, A., Belongie, S., & Shavit, N. (2018, February). Deep learning is robust to massive label noise. arXiv. https:\/\/doi.org\/10.48550\/arXiv.1705.10694"},{"key":"2025112614554962000_IMAG.a.1029-b27","first-page":"207","article-title":"Annotation-free deep-learning framework for microcalcifications detection on mammograms","volume":"12927","author":"Terrassin","year":"2024","journal-title":"Medical Imaging 2024: Computer-Aided Diagnosis"},{"issue":"2","key":"2025112614554962000_IMAG.a.1029-b28","doi-asserted-by":"crossref","first-page":"162","DOI":"10.1016\/S1474-4422(17)30470-2","article-title":"Diagnosis of multiple sclerosis: 2017 revisions of the McDonald criteria","volume":"17","author":"Thompson","year":"2018","journal-title":"The Lancet Neurology"},{"issue":"1","key":"2025112614554962000_IMAG.a.1029-b29","doi-asserted-by":"crossref","first-page":"118","DOI":"10.1177\/1352458518815602","article-title":"Observatoire Fran\u00e7ais de la Scl\u00e9rose en Plaques (OFSEP): A unique multimodal nationwide MS registry in France","volume":"26","author":"Vukusic","year":"2020","journal-title":"Multiple Sclerosis Journal"},{"key":"2025112614554962000_IMAG.a.1029-b30","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-72114-4_46","article-title":"Multi-sequence learning for multiple sclerosis lesion segmentation in spinal cord MRI","volume-title":"MICCAI","author":"Walsh","year":"2024"},{"key":"2025112614554962000_IMAG.a.1029-b31","doi-asserted-by":"crossref","first-page":"463","DOI":"10.1109\/CBMS58004.2023.00263","article-title":"Expert variability and deep learning performance in spinal cord lesion segmentation for multiple sclerosis patients","volume-title":"2023 IEEE 36th International Symposium on Computer-Based Medical Systems (CBMS)","author":"Walsh","year":"2023"},{"issue":"1","key":"2025112614554962000_IMAG.a.1029-b32","doi-asserted-by":"crossref","first-page":"67","DOI":"10.3390\/diagnostics13010067","article-title":"A comparison of techniques for class imbalance in deep learning classification of breast cancer","volume":"13","author":"Walsh","year":"2023","journal-title":"Diagnostics"},{"issue":"8","key":"2025112614554962000_IMAG.a.1029-b33","doi-asserted-by":"crossref","first-page":"653","DOI":"10.1016\/S1474-4422(21)00095-8","article-title":"2021 MAGNIMS\u2013CMSC\u2013NAIMS consensus recommendations on the use of MRI in patients with multiple sclerosis","volume":"20","author":"Wattjes","year":"2021","journal-title":"The Lancet Neurology"},{"key":"2025112614554962000_IMAG.a.1029-b34","first-page":"6022","article-title":"CutMix: Regularization strategy to train strong classifiers with localizable features","volume-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision","author":"Yun","year":"2019"},{"key":"2025112614554962000_IMAG.a.1029-b35","article-title":"Mixup: Beyond empirical risk minimization","author":"Zhang","year":"2018","journal-title":"International Conference on Learning Representations"},{"key":"2025112614554962000_IMAG.a.1029-b36","doi-asserted-by":"crossref","first-page":"120041","DOI":"10.1016\/j.neuroimage.2023.120041","article-title":"CarveMix: A simple data augmentation method for brain lesion segmentation","volume":"271","author":"Zhang","year":"2023","journal-title":"NeuroImage"},{"key":"2025112614554962000_IMAG.a.1029-b37","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-031-16440-8_65","article-title":"SelfMix: A self-adaptive data augmentation method for lesion segmentation","volume-title":"Medical Image Computing and Computer Assisted Intervention\u2014MICCAI 2022","author":"Zhu","year":"2022"}],"container-title":["Imaging Neuroscience"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/imag\/article-pdf\/doi\/10.1162\/IMAG.a.1029\/2563015\/imag.a.1029.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/imag\/article-pdf\/doi\/10.1162\/IMAG.a.1029\/2563015\/imag.a.1029.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,26]],"date-time":"2025-11-26T19:55:58Z","timestamp":1764186958000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/imag\/article\/doi\/10.1162\/IMAG.a.1029\/133911\/LesionSCynth-A-simple-parametric-lesion-synthesis"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":37,"URL":"https:\/\/doi.org\/10.1162\/imag.a.1029","relation":{},"ISSN":["2837-6056"],"issn-type":[{"value":"2837-6056","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025]]},"published":{"date-parts":[[2025]]},"article-number":"IMAG.a.1029"}}