{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T17:50:00Z","timestamp":1782150600519,"version":"3.54.5"},"reference-count":43,"publisher":"Public Library of Science (PLoS)","issue":"6","license":[{"start":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T00:00:00Z","timestamp":1782086400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Neurological disorders affect approximately 3 billion people globally, yet clinical trial success is often hindered by poorly selected outcome measures, impacting trial design, compliance, and interpretation. Over the past 25 years, Core Outcome Sets (COS) have emerged as standardized tools to enhance outcome selection, ensuring comparability across studies and reflecting the priorities of both researchers and patients. Despite the success of COS initiatives in other fields, their development in neurology remains limited, leaving many trialists without disease-specific guidance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>This study aimed to conduct an AI-assisted thematic synthesis of outcome domains from existing COS, with the goal of identifying structural patterns common to these sets and generating descriptive reference framework to inform future COS development.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>COS-Neuro was developed using AI-assisted thematic framework analysis, complemented by expert review. A modified five-step thematic analysis was conducted without pre-determined codes: 1. Dataset Gathering \u2013 Data was collected from the COMET database, and COS domains for neurological disorders were coded. 2. Prompt Design &amp; Testing \u2013 Large language models (LLMs), including ChatGPT 3.5, Google Gemini 1.5 Flash and Meta Llama-2-70b, were trialled, and prompts refined based on their outputs. 3. Thematic Analysis \u2013 LLMs categorised domains into core areas. 4. Human Refinement \u2013 Experts reviewed LLM-generated core areas and selected those most appropriate for further interpretation. 5. Clinical Validation \u2013 Experts validated the domains, core areas, and concepts. This approach integrated AI with expert oversight to develop an AI-descriptive thematic map of existing neurological COS.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec023\">\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Utilising LLMs, particularly ChatGPT, an AI-assisted conceptual framework synthesising existing neurological COS was developed based on the analysis of 112 published COS. Through adaptation of the OMERACT model, the final framework comprised four overarching concepts, 13 core areas, and 75 domains identified through expert consensus.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec024\">\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>COS-Neuro provides a preliminary AI-assisted descriptive synthesis of existing neurological COS, organised using the OMERACT Filter 2.1 as a structural reference. This hypothesis-generating framework may serve as a foundational resource for future COS research and trial design, particularly in areas where no disease-specific COS exists. However, it requires prospective validation through disease-specific, multi-stakeholder consensus processes before clinical application. COS-Neuro also demonstrates the feasibility of AI-assisted thematic synthesis in this context and warrants evaluation in other specialties.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pone.0333864","type":"journal-article","created":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T17:25:56Z","timestamp":1782149156000},"page":"e0333864","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["AI-assisted thematic synthesis of existing neurological core outcome sets: A descriptive reference framework (COS-Neuro)"],"prefix":"10.1371","volume":"21","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8004-397X","authenticated-orcid":true,"given":"Shyun Ping","family":"Tiong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9055-1230","authenticated-orcid":true,"given":"Xiaoyu","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9309-1885","authenticated-orcid":true,"given":"Alvaro Yanez","family":"Touzet","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher Paul","family":"Millward","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4396-4796","authenticated-orcid":true,"given":"Carl M.","family":"Zipser","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lindsay","family":"Tetreault","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6705-1115","authenticated-orcid":true,"given":"Ali","family":"Gharooni","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Benjamin M","family":"Davies","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,6,22]]},"reference":[{"issue":"4","key":"pone.0333864.ref001","doi-asserted-by":"crossref","first-page":"344","DOI":"10.1016\/S1474-4422(24)00038-3","article-title":"Global, regional, and national burden of disorders affecting the nervous system, 1990-2021: a systematic analysis for the Global Burden of Disease Study 2021","volume":"23","author":"Collaborators GBD 2021 Nervous System Disorders","year":"2024","journal-title":"Lancet Neurol"},{"key":"pone.0333864.ref002","doi-asserted-by":"crossref","first-page":"117579","DOI":"10.1016\/j.jns.2021.117579","article-title":"Neurology trial registrations on ClinicalTrials.gov between 2007 and 2018: A cross-sectional analysis of characteristics, early discontinuation, and results reporting","volume":"428","author":"BE Turner","year":"2021","journal-title":"J Neurol Sci"},{"issue":"10","key":"pone.0333864.ref003","doi-asserted-by":"crossref","DOI":"10.4088\/JCP.15f10377","article-title":"The primary outcome measure and its importance in clinical trials","volume":"76","author":"C Andrade","year":"2015","journal-title":"J Clin Psychiatry"},{"key":"pone.0333864.ref004","doi-asserted-by":"crossref","DOI":"10.1177\/2192568219832855","article-title":"RE-CODE DCM (REsearch Objectives and Common Data Elements for Degenerative Cervical Myelopathy): A Consensus Process to Improve Research Efficiency in DCM, Through Establishment of a Standardized Dataset for Clinical Research and the Definition of the Research Priorities","volume":"9","author":"BM Davies","year":"2019","journal-title":"Global Spine J"},{"issue":"4","key":"pone.0333864.ref005","doi-asserted-by":"crossref","first-page":"633","DOI":"10.3899\/jrheum.191225","article-title":"Spinal research \u2014 a field in need of standardization","volume":"47","author":"DZ Khan","year":"2020","journal-title":"J Rheumatol"},{"issue":"2","key":"pone.0333864.ref006","doi-asserted-by":"crossref","first-page":"193","DOI":"10.2217\/cer-2015-0007","article-title":"Improving the relevance and consistency of outcomes in comparative effectiveness research","volume":"5","author":"SR Tunis","year":"2016","journal-title":"J Comp Eff Res"},{"key":"pone.0333864.ref007","doi-asserted-by":"crossref","first-page":"280","DOI":"10.1186\/s13063-017-1978-4","article-title":"The COMET Handbook: version 1.0","volume":"18","author":"PR Williamson","year":"2017","journal-title":"Trials"},{"key":"pone.0333864.ref008","doi-asserted-by":"crossref","first-page":"152411","DOI":"10.1016\/j.semarthrit.2024.152411","article-title":"Strategies to promote implementation of core outcomes for medication adherence trials in rheumatology: A report from the OMERACT-Adherence Group","volume":"66","author":"AM Kordkheili","year":"2024","journal-title":"Semin Arthritis Rheum"},{"key":"pone.0333864.ref009","first-page":"404","article-title":"Challenges to Global Standardization of Outcome Measures","volume":"2021","author":"Z Liao","year":"2021","journal-title":"AMIA Jt Summits Transl Sci Proc"},{"issue":"7","key":"pone.0333864.ref010","doi-asserted-by":"crossref","first-page":"1048","DOI":"10.1093\/neuonc\/noac062","article-title":"Opportunities and challenges for the development of \u201ccore outcome sets\u201d in neuro-oncology","volume":"24","author":"CP Millward","year":"2022","journal-title":"Neuro Oncol"},{"key":"pone.0333864.ref011","doi-asserted-by":"crossref","DOI":"10.1177\/16094069231205789","article-title":"A Step-by-Step Process of Thematic Analysis to Develop a Conceptual Model in Qualitative Research","volume":"22","author":"M Naeem","year":"2023","journal-title":"Int J Qual Methods"},{"issue":"2","key":"pone.0333864.ref012","first-page":"560","article-title":"Thematic analysis through artificial intelligence (AI)","volume":"29","author":"P Christou","year":"2024","journal-title":"Qual Rep"},{"issue":"11","key":"pone.0333864.ref013","doi-asserted-by":"crossref","first-page":"7163","DOI":"10.1007\/s10115-024-02157-9","article-title":"Correction: An analysis of large language models: their impact and potential applications","volume":"66","author":"G Bharathi Mohan","year":"2024","journal-title":"Knowl Inf Syst"},{"issue":"8","key":"pone.0333864.ref014","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pcbi.1011319","article-title":"Ten quick tips for harnessing the power of ChatGPT in computational biology","volume":"19","author":"T Lubiana","year":"2023","journal-title":"PLoS Comput Biol"},{"key":"pone.0333864.ref015","doi-asserted-by":"crossref","DOI":"10.2196\/59641","article-title":"Large language models can enable inductive thematic analysis of a social media corpus in a single prompt: human validation study","volume":"4","author":"MS Deiner","year":"2024","journal-title":"JMIR Infodemiol"},{"key":"pone.0333864.ref016","doi-asserted-by":"crossref","first-page":"117","DOI":"10.1186\/1471-2288-13-117","article-title":"Using the framework method for the analysis of qualitative data in multi-disciplinary health research","volume":"13","author":"NK Gale","year":"2013","journal-title":"BMC Med Res Methodol"},{"key":"pone.0333864.ref017"},{"key":"pone.0333864.ref018"},{"issue":"8","key":"pone.0333864.ref019","doi-asserted-by":"crossref","first-page":"1021","DOI":"10.3899\/jrheum.181096","article-title":"OMERACT Filter 2.1: Elaboration of the Conceptual Framework for Outcome Measurement in Health Intervention Studies","volume":"46","author":"M Boers","year":"2019","journal-title":"J Rheumatol"},{"key":"pone.0333864.ref020","author":"J Crosley","year":"2021"},{"key":"pone.0333864.ref021"},{"key":"pone.0333864.ref022","article-title":"ChatGPT vs Gemini vs LLaMA on Multilingual Sentiment Analysis: arxiv","author":"A Buscemi","year":"2024"},{"key":"pone.0333864.ref023","author":"X Amatriain","year":"2024","journal-title":"Prompt design and engineering: introduction and advanced methods: arXiv"},{"key":"pone.0333864.ref024","author":"S Sankar","year":"2023"},{"issue":"4","key":"pone.0333864.ref025","doi-asserted-by":"crossref","first-page":"1893","DOI":"10.1007\/s11135-017-0574-8","article-title":"Saturation in qualitative research: exploring its conceptualization and operationalization","volume":"52","author":"B Saunders","year":"2018","journal-title":"Qual Quant"},{"key":"pone.0333864.ref026","year":"2018"},{"issue":"7","key":"pone.0333864.ref027","doi-asserted-by":"crossref","first-page":"745","DOI":"10.1016\/j.jclinepi.2013.11.013","article-title":"Developing core outcome measurement sets for clinical trials: OMERACT filter 2.0","volume":"67","author":"M Boers","year":"2014","journal-title":"J Clin Epidemiol"},{"key":"pone.0333864.ref028","doi-asserted-by":"crossref","first-page":"111277","DOI":"10.1016\/j.jclinepi.2024.111277","article-title":"Study found increasing use of core outcome sets in Cochrane systematic reviews and identified facilitators and barriers","volume":"169","author":"IJ Saldanha","year":"2024","journal-title":"J Clin Epidemiol"},{"key":"pone.0333864.ref029","article-title":"Harnessing ChatGPT for Thematic Analysis: Are We Ready?","volume":"26","author":"VV Lee","year":"2024","journal-title":"J Med Internet Res"},{"issue":"7227","key":"pone.0333864.ref030","doi-asserted-by":"crossref","first-page":"114","DOI":"10.1136\/bmj.320.7227.114","article-title":"Qualitative research in health care. Analysing qualitative data","volume":"320","author":"C Pope","year":"2000","journal-title":"BMJ"},{"issue":"1","key":"pone.0333864.ref031","doi-asserted-by":"crossref","first-page":"13","DOI":"10.4314\/mmj.v27i1.4","article-title":"The Implication of Using NVivo Software in Qualitative Data Analysis: Evidence-Based Reflections","volume":"27","author":"FC Zamawe","year":"2015","journal-title":"Malawi Med J"},{"key":"pone.0333864.ref032","article-title":"A Comprehensive Overview of Large Language Models","author":"Naveed HK, Asad, Qiu S, Saqib M, Anwar S, Usman M, et al","year":"2023","journal-title":"J Latex"},{"key":"pone.0333864.ref033","doi-asserted-by":"crossref","DOI":"10.2196\/54482","article-title":"Comparing the Efficacy and Efficiency of Human and Generative AI: Qualitative Thematic Analyses","volume":"3","author":"MR Prescott","year":"2024","journal-title":"JMIR AI"},{"issue":"4","key":"pone.0333864.ref034","first-page":"620","article-title":"A mixed-methods study comparing human-led and ChatGPT-driven qualitative analysis in medical education research","volume":"86","author":"T Kondo","year":"2024","journal-title":"Nagoya J Med Sci"},{"key":"pone.0333864.ref035","doi-asserted-by":"crossref","DOI":"10.2196\/64447","article-title":"Large Language Models for Thematic Summarization in Qualitative Health Care Research: Comparative Analysis of Model and Human Performance","volume":"4","author":"A Castellanos","year":"2025","journal-title":"JMIR AI"},{"key":"pone.0333864.ref036","volume-title":"Using ChatGPT for thematic analysis","author":"A Turoboc","year":"2024"},{"key":"pone.0333864.ref037","doi-asserted-by":"crossref","DOI":"10.2196\/55957","article-title":"Toward Clinical Generative AI: Conceptual Framework","volume":"3","author":"NL Bragazzi","year":"2024","journal-title":"JMIR AI"},{"issue":"10","key":"pone.0333864.ref038","doi-asserted-by":"crossref","first-page":"1551","DOI":"10.3899\/jrheum.161273","article-title":"Engaging Stakeholders and Promoting Uptake of OMERACT Core Outcome Instrument Sets","volume":"44","author":"SR Tunis","year":"2017","journal-title":"J Rheumatol"},{"issue":"8","key":"pone.0333864.ref039","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pmed.1004447","article-title":"A minimum data set-Core outcome set, core data elements, and core measurement set-For degenerative cervical myelopathy research (AO Spine RECODE DCM): A consensus study","volume":"21","author":"BM Davies","year":"2024","journal-title":"PLoS Med"},{"issue":"1","key":"pone.0333864.ref040","doi-asserted-by":"crossref","DOI":"10.1136\/bmjmed-2022-000284","article-title":"Core outcome sets in medical research","volume":"1","author":"JJ Kirkham","year":"2022","journal-title":"BMJ Med"},{"key":"pone.0333864.ref041","volume-title":"ChatGPT (GPT-3.5)","author":"OpenAI","year":"2023"},{"key":"pone.0333864.ref042","volume-title":"Gemini 1.5 Flash","author":"GoogleDeepMind","year":"2024"},{"key":"pone.0333864.ref043","volume-title":"LLaMA 2 (70B parameter model)","author":"MetaAI","year":"2023"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0333864","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,22]],"date-time":"2026-06-22T17:26:02Z","timestamp":1782149162000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0333864"}},"subtitle":[],"editor":[{"given":"Kewei","family":"Chen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,6,22]]},"references-count":43,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6,22]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0333864","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,22]]}}}