{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T18:47:12Z","timestamp":1775069232032,"version":"3.50.1"},"reference-count":54,"publisher":"Public Library of Science (PLoS)","issue":"4","license":[{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["EP\/T517926\/1"],"award-info":[{"award-number":["EP\/T517926\/1"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>\n                    Multiple Sclerosis (MS) is an inflammatory demyelinating disease of the central nervous system, typically exhibiting radiologically identifiable lesions within the brain and spinal cord, features key to both diagnosis and clinical disease monitoring. Manually identifying and segmenting lesions is both difficult and time consuming, thus optimising approaches to reliably automate segmentation is highly beneficial. The aim of this study was to assess the impact of data augmentation and manipulation on the accuracy of automated lesion segmentation using MRI scans from MS patients. Factors examined include MRI modalities in both isolation and combination, image augmentation, lesion size and size of testing set relative to training. The MICCAI 2016 MS dataset was used in this study, with U-Net chosen as the algorithmic method for segmentation. Each factor was optimised and then combined to maximise segmentation accuracy; the Dice metric was used as the focal metric to assess the efficacy of any given permutation of the setup. Statistical significance was assessed using the Mann\u2013Whitney U-test, with each permutation repeated five to ten times to ensure robustness. The best Dice score achieved using the testing and training dataset as outlined in the MICCAI 2016 challenge rubric was 0.59, approximately a 2% improvement against controls. To achieve this result whilst adhering to the training and testing distribution as defined in the dataset publication, the optimal imaging sequence was determined to be proton density. The augmentation conditions used included implementing an additional rotation of the dataset (doubling it in size) and excluding lesions\u2009&lt;\u200936.43 mm\n                    <jats:sup>3<\/jats:sup>\n                    in volume. The impact of data manipulation and augmentation was found to be statistically significant against controls using a Mann-Whitney U-test for lesion segmentation. An ancillary finding of this study was that there was no statistically significant difference between using one MRI modality for training and another for testing.\n                  <\/jats:p>","DOI":"10.1371\/journal.pdig.0001316","type":"journal-article","created":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T17:29:55Z","timestamp":1775064595000},"page":"e0001316","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Reliability of a convolutional neural network in segmenting multiple sclerosis lesions from MRI: Impact of data augmentation, image modality and tolerance with U-Net architecture"],"prefix":"10.1371","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-1005-603X","authenticated-orcid":true,"given":"Adam C.","family":"Szekely-Kohn","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5623-7491","authenticated-orcid":true,"given":"Marco","family":"Castellani","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2883-706X","authenticated-orcid":true,"given":"Luca","family":"Baronti","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zubair","family":"Ahmed","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"William G. K.","family":"Manifold","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Michael","family":"Douglas","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Daniel M.","family":"Espino","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"340","published-online":{"date-parts":[[2026,4,1]]},"reference":[{"issue":"14","key":"pdig.0001316.ref001","doi-asserted-by":"crossref","first-page":"1816","DOI":"10.1177\/1352458520970841","article-title":"Rising prevalence of multiple sclerosis worldwide: Insights from the Atlas of MS, third edition","volume":"26","author":"C Walton","year":"2020","journal-title":"Mult Scler"},{"issue":"5","key":"pdig.0001316.ref002","doi-asserted-by":"crossref","first-page":"689","DOI":"10.1084\/jem.61.5.689","article-title":"Encephalomyelitis accompanied by myelin destruction experimentally produced in monkeys","volume":"61","author":"TM Rivers","year":"1935","journal-title":"J Exp Med"},{"key":"pdig.0001316.ref003","doi-asserted-by":"crossref","first-page":"2705","DOI":"10.1093\/brain\/awh641","article-title":"Cortical demyelination and diffuse white matter injury in multiple sclerosis","volume":"128","author":"A Kutzelnigg","year":"2005","journal-title":"Brain"},{"key":"pdig.0001316.ref004","volume-title":"Untersuchungen \u00fcber die Gewebsver\u00e4nderungen bei der Multiplen Sklerose des Gehirns und R\u00fcckenmarks","author":"C Frommann","year":"1878"},{"issue":"7","key":"pdig.0001316.ref005","doi-asserted-by":"crossref","first-page":"1858","DOI":"10.1093\/brain\/awz144","article-title":"Assessment of lesions on magnetic resonance imaging in multiple sclerosis: practical guidelines","volume":"142","author":"M Filippi","year":"2019","journal-title":"Brain"},{"issue":"3","key":"pdig.0001316.ref006","doi-asserted-by":"crossref","first-page":"292","DOI":"10.1016\/S1474-4422(15)00393-2","article-title":"MRI criteria for the diagnosis of multiple sclerosis: MAGNIMS consensus guidelines","volume":"15","author":"M Filippi","year":"2016","journal-title":"Lancet Neurol"},{"issue":"2","key":"pdig.0001316.ref007","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":"AJ Thompson","year":"2018","journal-title":"Lancet Neurol"},{"issue":"12","key":"pdig.0001316.ref008","doi-asserted-by":"crossref","first-page":"3737","DOI":"10.1007\/s10439-021-02866-0","article-title":"Investigation of the Compressive Viscoelastic Properties of Brain Tissue Under Time and Frequency Dependent Loading Conditions","volume":"49","author":"W Li","year":"2021","journal-title":"Ann Biomed Eng"},{"issue":"3","key":"pdig.0001316.ref009","doi-asserted-by":"crossref","DOI":"10.1101\/cshperspect.a028936","article-title":"Multiple Sclerosis Pathology","volume":"8","author":"H Lassmann","year":"2018","journal-title":"Cold Spring Harb Perspect Med"},{"issue":"1","key":"pdig.0001316.ref010","doi-asserted-by":"crossref","first-page":"2078","DOI":"10.1038\/s41467-021-22265-2","article-title":"Identifying multiple sclerosis subtypes using unsupervised machine learning and MRI data","volume":"12","author":"A Eshaghi","year":"2021","journal-title":"Nat Commun"},{"issue":"1","key":"pdig.0001316.ref011","doi-asserted-by":"crossref","first-page":"241052","DOI":"10.1098\/rsos.241052","article-title":"Machine learning for refining interpretation of magnetic resonance imaging scans in the management of multiple sclerosis: a narrative review","volume":"12","author":"AC Szekely-Kohn","year":"2025","journal-title":"R Soc Open Sci"},{"key":"pdig.0001316.ref012","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1146\/annurev-bioeng-071516-044442","article-title":"Deep Learning in Medical Image Analysis","volume":"19","author":"D Shen","year":"2017","journal-title":"Annu Rev Biomed Eng"},{"key":"pdig.0001316.ref013","unstructured":"Beaumont J, Commowick O, Barillot C. Multiple sclerosis lesion segmentation using an automated multimodal graph cut. In: MSSEG Challenge Proceedings: Multiple Sclerosis Lesions Segmentation Challenge Using a Data Management and Processing Infrastructure, 2016. 1\u20138."},{"key":"pdig.0001316.ref014","unstructured":"Mahbod A, Wang C, Smedby O. Automatic multiple sclerosis lesion segmentation using hybrid artificial neural networks. In: MSSEG Challenge Proceedings: Multiple Sclerosis Lesions Segmentation Challenge Using a Data Management and Processing Infrastructure, 2016. 29\u201336."},{"key":"pdig.0001316.ref015","doi-asserted-by":"crossref","unstructured":"La Rosa F, Fartaria MJ, Kober T, Richiardi J, Granziera C, Thiran JP, et al. Shallow vs deep learning architectures for white matter lesion segmentation in the early stages of multiple sclerosis. In: Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries: 4th International Workshop, BrainLes 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 16, 2018, Revised Selected Papers, Part I, 2019. 142\u201351.","DOI":"10.1007\/978-3-030-11723-8_14"},{"key":"pdig.0001316.ref016","doi-asserted-by":"crossref","unstructured":"Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. In: International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015. 234\u201341.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"pdig.0001316.ref017","doi-asserted-by":"crossref","unstructured":"Zhou Z, Rahman Siddiquee MM, Tajbakhsh N, Liang J. Unet++: A nested u-net architecture for medical image segmentation. In: Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, Held in Conjunction with MICCAI 2018, Granada, Spain, September 20, 2018, Proceedings 4. Springer; 2018:3\u201311.","DOI":"10.1007\/978-3-030-00889-5_1"},{"issue":"2","key":"pdig.0001316.ref018","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":"F Isensee","year":"2021","journal-title":"Nat Methods"},{"key":"pdig.0001316.ref019","doi-asserted-by":"crossref","first-page":"104699","DOI":"10.1016\/j.compbiomed.2021.104699","article-title":"Sharp U-Net: Depthwise convolutional network for biomedical image segmentation","volume":"136","author":"H Zunair","year":"2021","journal-title":"Comput Biol Med"},{"key":"pdig.0001316.ref020","article-title":"Attention U-Net: Learning Where to Look for the Pancreas.","author":"O Oktay","year":"2018"},{"key":"pdig.0001316.ref021","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1016\/j.media.2018.09.005","article-title":"Automatic grading of prostate cancer in digitized histopathology images: Learning from multiple experts","volume":"50","author":"G Nir","year":"2018","journal-title":"Med Image Anal"},{"key":"pdig.0001316.ref022","article-title":"Segmentation and generation of magnetic resonance images by deep neural networks","author":"A Delplace","year":"2020","journal-title":"arXiv preprint"},{"key":"pdig.0001316.ref023","doi-asserted-by":"crossref","first-page":"102003","DOI":"10.1016\/j.nicl.2019.102003","article-title":"Uncovering convolutional neural network decisions for diagnosing multiple sclerosis on conventional MRI using layer-wise relevance propagation","volume":"24","author":"F Eitel","year":"2019","journal-title":"Neuroimage Clin"},{"key":"pdig.0001316.ref024","doi-asserted-by":"crossref","unstructured":"Kumar A, Murthy ON, Shrish, Ghosal P, Mukherjee A, Nandi D. A dense U-Net architecture for multiple sclerosis lesion segmentation. In: 2019. 662\u20137.","DOI":"10.1109\/TENCON.2019.8929615"},{"key":"pdig.0001316.ref025","doi-asserted-by":"crossref","unstructured":"Ghosal P, Prasad PKC, Nandi D. A light weighted deep learning framework for multiple sclerosis lesion segmentation. In: IEEE, 2019. 526\u201331.","DOI":"10.1109\/ICIIP47207.2019.8985674"},{"issue":"9","key":"pdig.0001316.ref026","doi-asserted-by":"crossref","first-page":"1459","DOI":"10.1007\/s11548-021-02327-y","article-title":"Multiple sclerosis lesion segmentation from brain MRI using U-Net based on wavelet pooling","volume":"16","author":"A Alijamaat","year":"2021","journal-title":"Int J Comput Assist Radiol Surg"},{"key":"pdig.0001316.ref027","doi-asserted-by":"crossref","first-page":"13650","DOI":"10.1038\/s41598-018-31911-7","article-title":"Objective evaluation of multiple sclerosis lesion segmentation using a data management and processing infrastructure","volume":"8","author":"O Commowick","year":"2018","journal-title":"Scientific Reports"},{"issue":"5","key":"pdig.0001316.ref028","doi-asserted-by":"crossref","first-page":"1487","DOI":"10.1002\/jmri.26959","article-title":"Deep-Learning-Based Neural Tissue Segmentation of MRI in Multiple Sclerosis: Effect of Training Set Size","volume":"51","author":"PA Narayana","year":"2020","journal-title":"J Magn Reson Imaging"},{"key":"pdig.0001316.ref029","unstructured":"Commowick O, Cervenansky F, Cotton F, Dojat M. MSSEG-2 challenge proceedings: Multiple sclerosis new lesions segmentation challenge using a data management and processing infrastructure. In: 2021. 1\u2013118."},{"key":"pdig.0001316.ref030","doi-asserted-by":"crossref","first-page":"106391","DOI":"10.1016\/j.compbiomed.2022.106391","article-title":"Data augmentation for medical imaging: A systematic literature review","volume":"152","author":"F Garcea","year":"2023","journal-title":"Comput Biol Med"},{"key":"pdig.0001316.ref031","doi-asserted-by":"crossref","first-page":"118589","DOI":"10.1016\/j.neuroimage.2021.118589","article-title":"Multiple sclerosis lesions segmentation from multiple experts: The MICCAI 2016 challenge dataset","volume":"244","author":"O Commowick","year":"2021","journal-title":"Neuroimage"},{"issue":"3","key":"pdig.0001316.ref032","doi-asserted-by":"crossref","first-page":"297","DOI":"10.2307\/1932409","article-title":"Measures of the Amount of Ecologic Association Between Species","volume":"26","author":"LR Dice","year":"1945","journal-title":"Ecology"},{"key":"pdig.0001316.ref033","article-title":"3D Segmentation in the Clinic: A Grand Challenge II: MS lesion segmentation","author":"M Styner","year":"2008","journal-title":"The MIDAS Journal"},{"key":"pdig.0001316.ref034","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":"A Carass","year":"2017","journal-title":"Neuroimage"},{"key":"pdig.0001316.ref035","doi-asserted-by":"crossref","DOI":"10.3389\/fict.2016.00025","article-title":"Shanoir: Applying the Software as a Service Distribution Model to Manage Brain Imaging Research Repositories","volume":"3","author":"C Barillot","year":"2016","journal-title":"Front ICT"},{"issue":"1","key":"pdig.0001316.ref036","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":"S Vukusic","year":"2020","journal-title":"Mult Scler"},{"issue":"4","key":"pdig.0001316.ref037","doi-asserted-by":"crossref","first-page":"250","DOI":"10.1016\/j.neurad.2020.01.083","article-title":"New OFSEP recommendations for MRI assessment of multiple sclerosis patients: Special consideration for gadolinium deposition and frequent acquisitions","volume":"47","author":"J-C Brisset","year":"2020","journal-title":"J Neuroradiol"},{"issue":"3","key":"pdig.0001316.ref038","doi-asserted-by":"crossref","first-page":"133","DOI":"10.1016\/j.neurad.2014.12.001","article-title":"OFSEP, a nationwide cohort of people with multiple sclerosis: Consensus minimal MRI protocol","volume":"42","author":"Imaging Working Group of the Observatoire Fran\u00e7ais de la Scl\u00e9rose en Plaques","year":"2015","journal-title":"J Neuroradiol"},{"issue":"8","key":"pdig.0001316.ref039","doi-asserted-by":"crossref","first-page":"671","DOI":"10.1136\/jnnp.55.8.671","article-title":"EDMUS, a European database for multiple sclerosis","volume":"55","author":"C Confavreux","year":"1992","journal-title":"J Neurol Neurosurg Psychiatry"},{"issue":"1","key":"pdig.0001316.ref040","first-page":"1","article-title":"An overview on Taguchi method","volume":"1","author":"SK Karna","year":"2012","journal-title":"International journal of engineering and mathematical sciences"},{"key":"pdig.0001316.ref041","unstructured":"Cardoso MJ, Li W, Brown R, Ma N, Kerfoot E, Wang Y. Monai: An open-source framework for deep learning in healthcare. 2022. https:\/\/arxiv.org\/abs\/2211.02701"},{"key":"pdig.0001316.ref042","unstructured":"Beaumont J, Commowick O, Barillot C. Automatic multiple sclerosis lesion segmentation from intensity-normalized multi-channel MRI. In: MSSEG Challenge Proceedings: Multiple Sclerosis Lesions Segmentation Challenge Using a Data Management and Processing Infrastructure, 2016. 9\u201316."},{"key":"pdig.0001316.ref043","unstructured":"Knight J, Khademi A. MS lesion segmentation using FLAIR MRI only. In: MSSEG Challenge Proceedings: Multiple Sclerosis Lesions Segmentation Challenge Using a Data Management and Processing Infrastructure, 2016. 21\u20138."},{"key":"pdig.0001316.ref044","unstructured":"Vera-Olmos F, Melero H, Malpica N. Random forest for multiple sclerosis lesion segmentation. In: Proceedings of the 1st MICCAI Challenge on Multiple Sclerosis Lesions Segmentation Challenge Using a Data Management and Processing Infrastructure-MICCAI-MSSEG, 2016. 81\u20136."},{"key":"pdig.0001316.ref045","first-page":"142","article-title":"Shallow vs deep learning architectures for white matter lesion segmentation in the early stages of multiple sclerosis","author":"FL Rosa","year":"2018"},{"key":"pdig.0001316.ref046","doi-asserted-by":"crossref","unstructured":"Shanmugam D, Blalock D, Balakrishnan G, Guttag J. Better Aggregation in Test-Time Augmentation. In: 2021 IEEE\/CVF International Conference on Computer Vision (ICCV), 2021. 1194\u2013203. https:\/\/doi.org\/10.1109\/iccv48922.2021.00125","DOI":"10.1109\/ICCV48922.2021.00125"},{"key":"pdig.0001316.ref047","doi-asserted-by":"crossref","first-page":"3501","DOI":"10.1007\/s00170-021-08589-9","article-title":"Artificial neural network tools for predicting the functional response of ultrafast laser textured\/structured surfaces","volume":"119","author":"L Baronti","year":"2022","journal-title":"Int J Adv Manuf Technol"},{"key":"pdig.0001316.ref048","first-page":"8780","article-title":"Diffusion models beat gans on image synthesis","volume":"34","author":"P Dhariwal","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"pdig.0001316.ref049","doi-asserted-by":"crossref","unstructured":"Isola P, Zhu JY, Zhou T, Efros AA. Image-to-image translation with conditional adversarial networks. IEEE Explore. 2016.","DOI":"10.1109\/CVPR.2017.632"},{"key":"pdig.0001316.ref050","doi-asserted-by":"crossref","unstructured":"Singh NK, Raza K. Medical image generation using generative adversarial networks: A review. Health informatics: A computational perspective in healthcare. 2021:77\u201396.","DOI":"10.1007\/978-981-15-9735-0_5"},{"issue":"3","key":"pdig.0001316.ref051","doi-asserted-by":"crossref","first-page":"69","DOI":"10.3390\/jimaging9030069","article-title":"GANs for Medical Image Synthesis: An Empirical Study","volume":"9","author":"Y Skandarani","year":"2023","journal-title":"J Imaging"},{"key":"pdig.0001316.ref052","doi-asserted-by":"crossref","first-page":"103735","DOI":"10.1016\/j.compbiomed.2020.103735","article-title":"Addressing class imbalance in deep learning for small lesion detection on medical images","volume":"120","author":"A Bria","year":"2020","journal-title":"Computers in Biology and Medicine"},{"issue":"1","key":"pdig.0001316.ref053","doi-asserted-by":"crossref","first-page":"63","DOI":"10.1109\/TKDE.2006.17","article-title":"Training cost-sensitive neural networks with methods addressing the class imbalance problem","volume":"18","author":"Zhou Zhi-Hua","year":"2006","journal-title":"IEEE Trans Knowl Data Eng"},{"key":"pdig.0001316.ref054","doi-asserted-by":"crossref","unstructured":"Gosain A, Sardana S. Handling class imbalance problem using oversampling techniques: A review. In: 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), 2017. 79\u201385. http:\/\/dx.doi.org\/10.1109\/icacci.2017.8125820","DOI":"10.1109\/ICACCI.2017.8125820"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001316","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T17:30:05Z","timestamp":1775064605000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001316"}},"subtitle":[],"editor":[{"given":"Dylan A","family":"Mordaunt","sequence":"first","affiliation":[],"role":[{"role":"editor","vocabulary":"crossref"}]}],"short-title":[],"issued":{"date-parts":[[2026,4,1]]},"references-count":54,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,4,1]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001316","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,1]]}}}