{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T11:15:36Z","timestamp":1793531736663,"version":"4.5.0"},"reference-count":41,"publisher":"Elsevier BV","issue":"2","license":[{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"funder":[{"DOI":"10.13039\/100014440","name":"Gobierno de Espana Ministerio de Ciencia e Innovacion","doi-asserted-by":"publisher","award":["PTQ2021-012120"],"award-info":[{"award-number":["PTQ2021-012120"]}],"id":[{"id":"10.13039\/100014440","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["clinicalkey.com","clinicalkey.com.au","clinicalkey.es","clinicalkey.fr","clinicalkey.jp","elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Seminars in Ultrasound, CT and MRI"],"published-print":{"date-parts":[[2026,4]]},"DOI":"10.1053\/j.sult.2026.01.003","type":"journal-article","created":{"date-parts":[[2026,1,30]],"date-time":"2026-01-30T17:07:33Z","timestamp":1769792853000},"page":"99-107","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":1,"title":["High Accuracy but Low Explainability: The Challenge of Explainable Artificial Intelligence in Multiple Sclerosis Assessment From Magnetic Resonance Imaging Radiology Reports"],"prefix":"10.1053","volume":"47","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7200-5028","authenticated-orcid":false,"given":"Teodoro","family":"Mart\u00edn-Noguerol","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pilar","family":"L\u00f3pez-\u00dabeda","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jorge","family":"Escartin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Suyash","family":"Mohan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Antonio","family":"Luna","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1053\/j.sult.2026.01.003_bib1","first-page":"14","article-title":"Multiple sclerosis: 2024 update","volume":"6","author":"Klotz","year":"2025","journal-title":"Free Neuropathol"},{"key":"10.1053\/j.sult.2026.01.003_bib2","doi-asserted-by":"crossref","first-page":"201","DOI":"10.1080\/14737175.2024.2304116","article-title":"An update on the role of magnetic resonance imaging in predicting and monitoring multiple sclerosis progression","volume":"24","author":"Ananthavarathan","year":"2024","journal-title":"Expert Rev\u00a0Neurother"},{"key":"10.1053\/j.sult.2026.01.003_bib3","first-page":"e000788","article-title":"Assessment of axonal injury in multiple sclerosis: Combined analysis of serum light-chain neurofilaments and diffusion tensor imaging","volume":"6","author":"Jalilian","year":"2024","journal-title":"BMJ Neurol\u00a0Open"},{"key":"10.1053\/j.sult.2026.01.003_bib4","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":"Lancet Neurol"},{"key":"10.1053\/j.sult.2026.01.003_bib5","first-page":"106658","article-title":"The 2024 McDonald criteria show comparable sensitivity to previous criteria for late-onset multiple sclerosis","volume":"103","author":"Tena-Cucala","year":"2025","journal-title":"Mult\u00a0Scler\u00a0Relat\u00a0Disord"},{"key":"10.1053\/j.sult.2026.01.003_bib6","first-page":"449","article-title":"MS-specific deep learning segmentation of deep gray matter","volume":"27","author":"Koopman","year":"2021","journal-title":"Mult\u00a0Scler\u00a0J"},{"key":"10.1053\/j.sult.2026.01.003_bib7","first-page":"1256","article-title":"Potential applications and impact of ChatGPT in radiology","volume":"31","author":"Bajaj","year":"2024","journal-title":"Acad\u00a0Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib8","doi-asserted-by":"crossref","first-page":"373","DOI":"10.2214\/AJR.23.29198","article-title":"Potential use cases for ChatGPT in radiology reporting","volume":"221","author":"Elkassem","year":"2023","journal-title":"AJR Am\u00a0J\u00a0Roentgenol"},{"key":"10.1053\/j.sult.2026.01.003_bib9","first-page":"1","article-title":"Radiology in the era of large language models: The near and the dark side of the moon","volume":"33","author":"L\u00f3pez-\u00dabeda","year":"2023","journal-title":"Eur Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib10","first-page":"1239","article-title":"Strengths, weaknesses, opportunities, and threats analysis of artificial intelligence and machine learning applications in radiology","volume":"16","author":"Mart\u00edn-Noguerol","year":"2019","journal-title":"J\u00a0Am\u00a0Coll\u00a0Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib11","unstructured":"Mavrepis P, Makridis G, Fatouros G, et al: XAI for all: Can large language models simplify explainable AI?, arXiv preprint arXiv:2401.13110, 2024. https:\/\/doi.org\/10.48550\/arXiv.2401.13110."},{"key":"10.1053\/j.sult.2026.01.003_bib12","doi-asserted-by":"crossref","first-page":"82","DOI":"10.1016\/j.inffus.2019.12.012","article-title":"Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI","volume":"58","author":"Arrieta","year":"2019","journal-title":"Inf Fusion"},{"key":"10.1053\/j.sult.2026.01.003_bib13","first-page":"8","article-title":"Radiology, explicability and AI: Closing the gap","volume":"33","author":"L\u00f3pez-\u00dabeda","year":"2023","journal-title":"Eur\u00a0Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib14","first-page":"1271","article-title":"Natural language processing in radiology: update on clinical applications","author":"L\u00f3pez-\u00dabeda","year":"2022","journal-title":"J\u00a0Am\u00a0Coll\u00a0Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib15","first-page":"S189","article-title":"Ease of adoption of clinical natural language processing software: An evaluation of five systems","volume":"58","author":"Zheng","year":"2015","journal-title":"J Biomed\u00a0Inf"},{"key":"10.1053\/j.sult.2026.01.003_bib16","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1007\/s10278-017-0030-2","article-title":"Automated radiology report summarization using an open-source natural language processing pipeline","volume":"31","author":"Goff","year":"2018","journal-title":"J\u00a0Digit Imaging"},{"key":"10.1053\/j.sult.2026.01.003_bib17","first-page":"104066","article-title":"COVID-19 detection in radiological text reports integrating entity recognition","volume":"127","author":"L\u00f3pez-\u00dabeda","year":"2020","journal-title":"Comput Biol\u00a0Med"},{"key":"10.1053\/j.sult.2026.01.003_bib18","doi-asserted-by":"crossref","first-page":"2113","DOI":"10.1007\/s00330-023-10202-4","article-title":"Natural language processing deep learning models for the differential between high-grade gliomas and metastasis: What if the key is how we report them?","author":"Mart\u00edn-Noguerol","year":"2023","journal-title":"Eur Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib19","first-page":"247","article-title":"MRI in the assessment and monitoring of multiple sclerosis: An update on best practice","volume":"10","author":"Kaunzner","year":"2017","journal-title":"Ther\u00a0Adv\u00a0Neurol\u00a0Disord"},{"key":"10.1053\/j.sult.2026.01.003_bib20","doi-asserted-by":"crossref","first-page":"895","DOI":"10.1080\/14737175.2021.1957832","article-title":"Beyond McDonald: updated perspectives on MRI diagnosis of multiple sclerosis","volume":"21","author":"Rovira","year":"2021","journal-title":"Expert Rev\u00a0Neurother"},{"key":"10.1053\/j.sult.2026.01.003_bib21","first-page":"e1059","article-title":"Prognostic accuracy of NEDA-3 in long-term outcomes of multiple sclerosis","volume":"8","author":"Prosperini","year":"2021","journal-title":"Neurol\u00a0Neuroimmunol\u00a0Neuroinflamm"},{"key":"10.1053\/j.sult.2026.01.003_bib22","doi-asserted-by":"crossref","first-page":"2742","DOI":"10.1093\/brain\/awaa231","article-title":"Delay from treatment start to full effect of immunotherapies for multiple sclerosis","volume":"143","author":"Roos","year":"2020","journal-title":"Brain"},{"key":"10.1053\/j.sult.2026.01.003_bib23","doi-asserted-by":"crossref","unstructured":"Ribeiro MT, Singh S, Guestrin C: \u201cWhy should I trust you?\u201d Explaining the predictions of any classifier. Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. NAACL-HLT 2016, pp\u00a097-101, 2016. https:\/\/doi.org\/10.18653\/v1\/n16-3020.","DOI":"10.18653\/v1\/N16-3020"},{"key":"10.1053\/j.sult.2026.01.003_bib24","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":"Lancet Neurol"},{"key":"10.1053\/j.sult.2026.01.003_bib25","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":"Filippi","year":"2019","journal-title":"Brain"},{"key":"10.1053\/j.sult.2026.01.003_bib26","doi-asserted-by":"crossref","first-page":"1557","DOI":"10.1016\/j.ejrnm.2016.07.014","article-title":"One step diagnosis of multiple sclerosis disease activity, dissemination in time and space using diffusion weighted MRI","volume":"47","author":"Radwan","year":"2016","journal-title":"Egypt J Radiol Nucl\u00a0Med"},{"key":"10.1053\/j.sult.2026.01.003_bib27","series-title":"Advances in Neural Information Processing Systems","first-page":"5999","article-title":"Attention is all you need","author":"Vaswani","year":"2017"},{"key":"10.1053\/j.sult.2026.01.003_bib28","first-page":"e210258","article-title":"RadBERT: Adapting transformer-based language models to radiology","volume":"4","author":"Yan","year":"2022","journal-title":"Radiol Artif\u00a0Intell"},{"key":"10.1053\/j.sult.2026.01.003_bib29","unstructured":"Carrino CP, Armengol-Estap\u00e9 J, Guti\u00e9rrez-Fandi\u00f1o A, et al: Biomedical and clinical language models for Spanish: On the benefits of domain-specific pretraining in a mid-resource scenario, arXiv preprint arXiv:2109.03570, 2021, https:\/\/doi.org\/10.48550\/arXiv.2109.03570."},{"key":"10.1053\/j.sult.2026.01.003_bib30","doi-asserted-by":"crossref","first-page":"113647","DOI":"10.1016\/j.eswa.2020.113647","article-title":"Detection of unexpected findings in radiology reports: A comparative study of machine learning approaches","volume":"160","author":"L\u00f3pez-\u00dabeda","year":"2020","journal-title":"Expert Syst\u00a0Appl"},{"key":"10.1053\/j.sult.2026.01.003_bib31","first-page":"1755","article-title":"Qualifying certainty in radiology reports through deep learning-based natural language processing","volume":"42","author":"Liu","year":"2021","journal-title":"Am\u00a0J\u00a0Neuroradiol"},{"key":"10.1053\/j.sult.2026.01.003_bib32","doi-asserted-by":"crossref","first-page":"131","DOI":"10.1007\/s10278-019-00271-7","article-title":"Automated detection of radiology reports that require follow-up imaging using natural language processing feature engineering and machine learning classification","volume":"33","author":"Lou","year":"2020","journal-title":"J\u00a0Digit Imaging"},{"key":"10.1053\/j.sult.2026.01.003_bib33","first-page":"111638","article-title":"AI supported detection of cerebral multiple sclerosis lesions decreases radiologic reporting times","volume":"178","author":"Peters","year":"2024","journal-title":"Eur\u00a0J\u00a0Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib34","first-page":"106620","article-title":"Automated multiple sclerosis lesion segmentation from 3D-FLAIR MRI using R2AUNet: A deep learning approach with recurrent residual and attention mechanisms","volume":"102","author":"Andishgar","year":"2025","journal-title":"Mult\u00a0Scler\u00a0Relat\u00a0Disord"},{"key":"10.1053\/j.sult.2026.01.003_bib35","first-page":"e394","article-title":"Machine learning-based classification of multiple sclerosis lesion activity using multi-sequence MRI radiomics: A\u00a0complete analysis of T1, T2, FLAIR, DWI, and SWI features","volume":"90","author":"Elhaie","year":"2025","journal-title":"Pol J\u00a0Radiol"},{"key":"10.1053\/j.sult.2026.01.003_bib36","first-page":"101860","article-title":"Sequence-level machine learning assessment of MRI protocols in classifying multiple sclerosis lesions","volume":"18","author":"Elhaie","year":"2025","journal-title":"J\u00a0Radiat\u00a0Res Appl\u00a0Sci"},{"key":"10.1053\/j.sult.2026.01.003_bib37","doi-asserted-by":"crossref","first-page":"3637","DOI":"10.1007\/s00234-025-03802-y","article-title":"A multi-center study of transformer-based CNNs for multiple sclerosis lesion segmentation on 3D FLAIR MRI","author":"Saadh","year":"2025","journal-title":"Neuroradiology"},{"key":"10.1053\/j.sult.2026.01.003_bib38","doi-asserted-by":"crossref","first-page":"1167612","DOI":"10.3389\/fnins.2023.1167612","article-title":"Multiple sclerosis lesion segmentation: Revisiting weighting mechanisms for federated learning","volume":"17","author":"Liu","year":"2023","journal-title":"Front Neurosci"},{"key":"10.1053\/j.sult.2026.01.003_bib39","first-page":"233","article-title":"Artificial intelligence and multiple sclerosis","volume":"24","author":"Amin","year":"2024","journal-title":"Curr\u00a0Neurol\u00a0Neurosci\u00a0Rep"},{"key":"10.1053\/j.sult.2026.01.003_bib40","doi-asserted-by":"crossref","first-page":"103611","DOI":"10.1016\/j.nicl.2024.103611","article-title":"LST-AI: A deep learning ensemble for accurate MS lesion segmentation","volume":"42","author":"Wiltgen","year":"2024","journal-title":"NeuroImage Clin"},{"key":"10.1053\/j.sult.2026.01.003_bib41","first-page":"13","article-title":"Augmented intelligence for advancing healthcare","volume":"20","author":"Ilyas","year":"2022","journal-title":"J Syst\u00a0Cyber Inform"}],"container-title":["Seminars in Ultrasound, CT and MRI"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S088721712600003X?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S088721712600003X?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T22:01:38Z","timestamp":1774044098000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S088721712600003X"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":41,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,4]]}},"alternative-id":["S088721712600003X"],"URL":"https:\/\/doi.org\/10.1053\/j.sult.2026.01.003","relation":{},"ISSN":["0887-2171"],"issn-type":[{"value":"0887-2171","type":"print"}],"subject":[],"published":{"date-parts":[[2026,4]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"High Accuracy but Low Explainability: The Challenge of Explainable Artificial Intelligence in Multiple Sclerosis Assessment From Magnetic Resonance Imaging Radiology Reports","name":"articletitle","label":"Article Title"},{"value":"Seminars in Ultrasound, CT and MRI","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1053\/j.sult.2026.01.003","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier Inc. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}]}}