{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T09:06:43Z","timestamp":1784192803997,"version":"3.55.0"},"reference-count":36,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T00:00:00Z","timestamp":1784160000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Brain Sciences"],"abstract":"<jats:p>Background: Multiple sclerosis (MS) lacks a single invariant phenotypic core. Patients accumulate heterogeneous combinations of sensory, motor, cognitive, and autonomic impairments over time, reflecting lesions that are disseminated in time and space. Standard scales such as the Expanded Disability Status Scale (EDSS) distribute disability across functional systems, but do not explicitly represent MS phenotype as a mixture of latent symptom modules. Methods: We analyzed 4617 de-identified neurology progress notes from 577 patients with MS at a single academic medical center. A large language model (GPT-5.2) categorized each note with respect to 17 non-mutually-exclusive neurological phenotype features, and note-level features were aggregated to patient-level binary vectors. Non-negative matrix factorization (NMF) was applied to generate three-, four-, and five-module solutions. For each rank, we computed approximate variance captured, relative reconstruction error, and module-level feature loadings. In the preferred four-module solution, we derived patient-level module percentages, identified highly dominant (\u226555%) and archetypal (\u226570%) module profiles, and quantified admixture using Shannon entropy and the effective number of modules. Results: Three-, four-, and five-module NMF solutions showed similar approximate variance captured (52.7\u201354.3%) and reconstruction error (0.47\u20130.53), but the four-module solution provided the clearest clinical interpretation. The four latent modules were sensory-visual-pain, ataxic-spastic-falls, cognitive-psychologic-fatigue, and autonomic-bladder-bowel, aligning closely with established functional systems in MS. Most patients exhibited admixed phenotypes, with module entropies ranging from 0 (single-module dominance) to 1.386 (equal mixture) and effective modules spanning approximately 1 to 4. Using pre-specified thresholds, 154 patients (26.6%) were highly dominant in a single module and 72 (12.5%) were archetypal; these purer phenotypes were most often in the sensory-visual-pain module. Conclusions: MS phenotypic diversity in routine clinical practice can be parsimoniously represented as mixtures of four latent symptom modules rather than as positions along a single severity axis. Most patients show substantial admixture of sensory, motor, cognitive, and autonomic involvement, but a minority exhibit relatively pure or strongly dominant module patterns. This modular representation provides an interpretable framework for quantifying MS phenotype and for generating testable hypotheses about MS subtypes whose biological relevance remains to be established.<\/jats:p>","DOI":"10.3390\/brainsci16070753","type":"journal-article","created":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T08:08:34Z","timestamp":1784189314000},"page":"753","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Phenotypic Diversity in Multiple Sclerosis Can Be Represented by Four Additive Symptom Modules"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6179-0793","authenticated-orcid":false,"given":"Daniel B.","family":"Hier","sequence":"first","affiliation":[{"name":"Department of Neurology and Rehabilitation, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-0336-2360","authenticated-orcid":false,"given":"Pavankumar Y.","family":"Srinivasula","sequence":"additional","affiliation":[{"name":"College of Liberal Arts and Sciences, University of Illinois Chicago, Chicago, IL 60607, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5337-0804","authenticated-orcid":false,"given":"Michael D.","family":"Carrithers","sequence":"additional","affiliation":[{"name":"Department of Neurology and Rehabilitation, College of Medicine, University of Illinois Chicago, Chicago, IL 60612, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,16]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"e1494","DOI":"10.1002\/widm.1494","article-title":"Unsupervised EHR-based phenotyping via matrix and tensor decompositions","volume":"13","author":"Becker","year":"2023","journal-title":"Wiley Interdiscip. Rev. Data Min. Knowl. Discov."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1212\/01.CON.0000293634.15851.7d","article-title":"Clinical features of multiple sclerosis","volume":"10","author":"Miller","year":"2004","journal-title":"Contin. Lifelong Learn. Neurol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"380","DOI":"10.7861\/clinmed.2020-0292","article-title":"Clinical presentation and diagnosis of multiple sclerosis","volume":"20","author":"Ford","year":"2020","journal-title":"Clin. Med."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"311","DOI":"10.34218\/IJCET_14_03_029","article-title":"Detection of Dementia using Gradient Boosting","volume":"14","author":"Bakhshizada","year":"2023","journal-title":"Int. J. Comput. Eng. Technol. (IJCET)"},{"key":"ref_5","first-page":"66","article-title":"Brain Tumor Classification using EfficientNet","volume":"187","author":"Billah","year":"2026","journal-title":"Int. J. Comput. Appl."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"13653","DOI":"10.1038\/s41598-022-17741-8","article-title":"Multiple sclerosis diagnosis and phenotype identification by multivariate classification of in vivo frontal cortex metabolite profiles","volume":"12","author":"Swanberg","year":"2022","journal-title":"Sci. Rep."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Correia, I., Bernardes, C., Cunha, C., Nunes, C., Mac\u00e1rio, C., Sousa, L., and Batista, S. (2024). Picturing the Multiple Sclerosis patient journey: A symptomatic overview. J. Clin. Med., 13.","DOI":"10.3390\/jcm13195687"},{"key":"ref_8","first-page":"203183","article-title":"Symptoms and association with health outcomes in relapsing-remitting multiple sclerosis: Results of a US patient survey","volume":"2014","author":"Williams","year":"2014","journal-title":"Mult. Scler. Int."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1111\/j.1600-0404.1970.tb05808.x","article-title":"Neurologic impairment in multiple sclerosis and the disability status scale","volume":"46","author":"Kurtzke","year":"1970","journal-title":"Acta Neurol. Scand."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1444","DOI":"10.1212\/WNL.33.11.1444","article-title":"Rating neurologic impairment in multiple sclerosis: An expanded disability status scale (EDSS)","volume":"33","author":"Kurtzke","year":"1983","journal-title":"Neurology"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"1027","DOI":"10.1093\/brain\/123.5.1027","article-title":"Kurtzke scales revisited: The application of psychometric methods to clinical intuition","volume":"123","author":"Hobart","year":"2000","journal-title":"Brain"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1336","DOI":"10.1109\/TKDE.2012.51","article-title":"Nonnegative matrix factorization: A comprehensive review","volume":"25","author":"Wang","year":"2012","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"788","DOI":"10.1038\/44565","article-title":"Learning the parts of objects by non-negative matrix factorization","volume":"401","author":"Lee","year":"1999","journal-title":"Nature"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"207","DOI":"10.1016\/j.neuroimage.2013.12.015","article-title":"Non-negative matrix factorization of multimodal MRI, fMRI and phenotypic data reveals differential changes in default mode subnetworks in ADHD","volume":"102","author":"Anderson","year":"2014","journal-title":"NeuroImage"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"207","DOI":"10.1016\/0024-3795(84)90096-X","article-title":"The nonnegative rank factorizations of nonnegative matrices","volume":"62","author":"Chen","year":"1984","journal-title":"Linear Algebra Its Appl."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Hier, D.B., Srinivasula, P.Y., and Carrithers, M.D. (2026). Note-Level Phenotyping of Multiple Sclerosis Notes by a Large Language Model Achieves Near Human-Level Agreement. J. Clin. Med., 15.","DOI":"10.20944\/preprints202604.1173.v1"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"176","DOI":"10.1016\/j.jbi.2004.11.009","article-title":"Primary care physician time utilization before and after implementation of an electronic health record: A time-motion study","volume":"38","author":"Pizziferri","year":"2005","journal-title":"J. Biomed. Inform."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"657","DOI":"10.1016\/j.ijmedinf.2005.03.006","article-title":"Differing faculty and housestaff acceptance of an electronic health record","volume":"74","author":"Hier","year":"2005","journal-title":"Int. J. Med. Inform."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"103821","DOI":"10.1016\/j.msard.2022.103821","article-title":"Feedback from neurology residents and neuroimmunology fellows about the practical training in multiple sclerosis","volume":"63","author":"Silva","year":"2022","journal-title":"Mult. Scler. Relat. Disord."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"81","DOI":"10.1109\/LSP.2019.2957604","article-title":"Near-Convex Archetypal Analysis","volume":"27","author":"Handschutter","year":"2019","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_21","first-page":"1","article-title":"Sparse nmf with archetypal regularization: Computational and robustness properties","volume":"25","author":"Behdin","year":"2024","journal-title":"J. Mach. Learn. Res."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"896","DOI":"10.1080\/01621459.2019.1594832","article-title":"Nonnegative Matrix Factorization Via Archetypal Analysis","volume":"115","author":"Javadi","year":"2017","journal-title":"J. Am. Stat. Assoc."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1080\/00401706.2016.1247017","article-title":"A Geometric Approach to Archetypal Analysis and Nonnegative Matrix Factorization","volume":"59","author":"Damle","year":"2014","journal-title":"Technometrics"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1109\/MSP.2018.2877582","article-title":"Nonnegative matrix factorization for signal and data analytics: Identifiability, algorithms, and applications","volume":"36","author":"Fu","year":"2019","journal-title":"IEEE Signal Process. Mag."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"552","DOI":"10.1111\/j.1749-6632.1965.tb20235.x","article-title":"Problems of experimental trials of therapy in multiple sclerosis: Report by the panel on the evaluation of experimental trials of therapy in multiple sclerosis","volume":"122","author":"Schumacher","year":"1965","journal-title":"Ann. N. Y. Acad. Sci."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"580","DOI":"10.1212\/WNL.5.8.580","article-title":"A new scale for evaluating disability in multiple sclerosis","volume":"5","author":"Kurtzke","year":"1955","journal-title":"Neurology"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"7553","DOI":"10.1109\/JBHI.2024.3453602","article-title":"Dynamite: Integrating archetypal analysis and process mining for interpretable disease progression modelling","volume":"28","author":"Trescato","year":"2024","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"617","DOI":"10.1007\/s11136-014-0804-7","article-title":"Contribution of symptom clusters to multiple sclerosis consequences","volume":"24","author":"Shahrbanian","year":"2015","journal-title":"Qual. Life Res."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1097\/00006199-198905000-00012","article-title":"Model confirmation of the MS-related symptom checklist","volume":"38","author":"Gulick","year":"1989","journal-title":"Nurs. Res."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Ajdacic-Gross, V., Steinemann, N., Horv\u00e1th, G., Rodgers, S., Kaufmann, M., Xu, Y., Kamm, C.P., Kesselring, J., Manjaly, Z.M., and Zecca, C. (2021). Onset symptom clusters in multiple sclerosis: Characteristics, comorbidities, and risk factors. Front. Neurol., 12.","DOI":"10.3389\/fneur.2021.693440"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1642","DOI":"10.1177\/13524585241282763","article-title":"Multiple sclerosis subgroups: Data-driven clusters based on patient-reported outcomes and a large clinical sample","volume":"30","author":"Zamora","year":"2024","journal-title":"Mult. Scler. J."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Howlett-Prieto, Q., Oommen, C., Carrithers, M.D., Wunsch, D.C., and Hier, D.B. (2023). Subtypes of relapsing-remitting multiple sclerosis identified by network analysis. Front. Digit. Health, 4.","DOI":"10.3389\/fdgth.2022.1063264"},{"key":"ref_33","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":"Eshaghi","year":"2021","journal-title":"Nat. Commun."},{"key":"ref_34","unstructured":"Herndon, R.M. (1997). Handbook of Neurologic Rating Scales, Demos Medical Publishing."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1038\/s41582-020-00448-7","article-title":"Standardization and digitization of clinical data in multiple sclerosis","volume":"17","author":"Papadopoulou","year":"2021","journal-title":"Nat. Rev. Neurol."},{"key":"ref_36","first-page":"100891","article-title":"Consensus quality indicators for monitoring multiple sclerosis","volume":"40","author":"Voigt","year":"2024","journal-title":"Lancet Reg. Health\u2013Eur."}],"container-title":["Brain Sciences"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2076-3425\/16\/7\/753\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,16]],"date-time":"2026-07-16T08:41:40Z","timestamp":1784191300000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2076-3425\/16\/7\/753"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,16]]},"references-count":36,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["brainsci16070753"],"URL":"https:\/\/doi.org\/10.3390\/brainsci16070753","relation":{},"ISSN":["2076-3425"],"issn-type":[{"value":"2076-3425","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,16]]}}}