{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T08:41:07Z","timestamp":1788511267151,"version":"build-2803163510"},"reference-count":40,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T00:00:00Z","timestamp":1785283200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Public Health"],"abstract":"<jats:p>Urgent and emergency care generates large volumes of clinical data, yet much of the information most useful for research, surveillance, and service planning is recorded only in free-text clinical notes and never reaches the structured datasets used for analysis. As a result, the reasons patients present, the upstream pressures driving demand, and the care that was needed but unavailable remain largely invisible, with consequences that fall most heavily on people who move repeatedly across services. Existing initiatives connect previously separate records or predict future events, but neither recovers the clinical reasoning that explains why people reach crisis. In this Perspective we propose a three-layer knowledge graph architecture for urgent and emergency care: a first layer linking individual contacts into connected patient trajectories across services and time; a second recovering clinical detail lost to reductive coding; and a third capturing the reasoning behind each presentation, including barriers to care and intended but unavailable services. We describe why recent advances in natural language processing and artificial intelligence make this architecture achievable, assess the current evidence - strong for each component, though the integrated architecture awaits validation - and set out the validation, governance, and infrastructure that implementation would require. The architecture offers a route from data volume to usable knowledge, and a means of making patterns of unmet need legible to those who plan and govern care. The constituent technologies now exist; whether this information remains invisible is increasingly a question of design rather than capability.<\/jats:p>","DOI":"10.3389\/fpubh.2026.1901339","type":"journal-article","created":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T06:03:32Z","timestamp":1785305012000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Natural language processing-driven knowledge graphs for transformative public health intelligence and research datasets in urgent and emergency care"],"prefix":"10.3389","volume":"14","author":[{"given":"Chris","family":"Humphries","sequence":"first","affiliation":[{"name":"Generative AI Laboratory, School of Informatics, University of Edinburgh","place":["Edinburgh, United Kingdom"]},{"name":"Centre for Cardiovascular Science, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Atul","family":"Anand","sequence":"additional","affiliation":[{"name":"Centre for Cardiovascular Science, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arlene","family":"Casey","sequence":"additional","affiliation":[{"name":"Centre for Population Health Sciences, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wendy","family":"Chapman","sequence":"additional","affiliation":[{"name":"Clinical Informatics Center, University of Texas Southwestern Medical Center","place":["Dallas, TX, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kathy","family":"Harrison","sequence":"additional","affiliation":[{"name":"Centre for Population Health Sciences, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"James M.","family":"Hillis","sequence":"additional","affiliation":[{"name":"Harvard Medical School","place":["Boston, MA, United States"]},{"name":"Mass General Brigham AI and Department of Neurology, Mass General Brigham","place":["Boston, MA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Edward","family":"James","sequence":"additional","affiliation":[{"name":"Department of Emergency Medicine, NHS Lothian","place":["Edinburgh, United Kingdom"]},{"name":"Centre for Sustainable Delivery, Golden Jubilee National Hospital","place":["Glasgow, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Donald J.","family":"MacIntyre","sequence":"additional","affiliation":[{"name":"Generative AI Laboratory, School of Informatics, University of Edinburgh","place":["Edinburgh, United Kingdom"]},{"name":"Centre for Cardiovascular Science, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kenneth","family":"Nicholson","sequence":"additional","affiliation":[{"name":"Public Health Scotland","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lisa","family":"Sch\u00f6lin","sequence":"additional","affiliation":[{"name":"Centre for Cardiovascular Science, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"R. Andrew","family":"Taylor","sequence":"additional","affiliation":[{"name":"Department of Emergency Medicine, University of Virginia","place":["Charlottesville, VA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher","family":"Tomlinson","sequence":"additional","affiliation":[{"name":"Department of Biostatistics & Health Informatics, King\u2019s College London","place":["London, United Kingdom"]},{"name":"Institute of Health Informatics, University College London","place":["London, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Robertson","sequence":"additional","affiliation":[{"name":"School of Informatics, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,29]]},"reference":[{"key":"ref1","article-title":"Care of vulnerable patients and those likely to experience health inequalities in the emergency department","year":"2025"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"628","DOI":"10.1136\/emermed-2024-214155","article-title":"At the deep end: towards a social emergency medicine","volume":"41","author":"McHenry","year":"2024","journal-title":"Emerg Med J"},{"key":"ref3","volume-title":"Health Equity in England: the Marmot Review 10 Years on Institute of Health Equity","author":"Marmot","year":"2020"},{"key":"ref4","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1016\/S0140-6736(17)31869-X","article-title":"Morbidity and mortality in homeless individuals, prisoners, sex workers, and individuals with substance use disorders in high-income countries: a systematic review and meta-analysis","volume":"391","author":"Aldridge","year":"2018","journal-title":"Lancet"},{"key":"ref5","article-title":"Hard edges: mapping severe and multiple disadvantage in England","year":"2015"},{"key":"ref6","article-title":"FRAIL strategy: a strategy for the development and\/or improvement of acute frailty same day emergency care services","year":"2024"},{"key":"ref7","first-page":"409","article-title":"All health care is not local: an evaluation of the distribution of emergency department care delivered in Indiana","volume":"2011","author":"Finnell","year":"2011","journal-title":"AMIA Annu Symp Proc"},{"key":"ref8","doi-asserted-by":"publisher","first-page":"1104","DOI":"10.23889\/ijpds.v4i1.1104","article-title":"Using deterministic record linkage to link ambulance and emergency department data: is it possible without patient identifiers? A case study from the UK","volume":"4","author":"Clark","year":"2019","journal-title":"Int J Popul Data Sci"},{"key":"ref9","doi-asserted-by":"publisher","first-page":"566","DOI":"10.1080\/10903127.2018.1549628","article-title":"Outcomes for patients who contact the emergency ambulance service and are not transported to the emergency department: a data linkage study","volume":"23","author":"Coster","year":"2019","journal-title":"Prehosp Emerg Care"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"e716","DOI":"10.1016\/s2468-2667(25)00096-9","article-title":"De-siloing substance misuse and self-harm research through integrated public health and emergency medicine","volume":"10","author":"Sch\u00f6lin","year":"2025","journal-title":"Lancet Public Health"},{"key":"ref11","doi-asserted-by":"publisher","first-page":"emermed-2025-215530","DOI":"10.1136\/emermed-2025-215530","article-title":"Unlocking clinical narratives: how natural language processing and artificial intelligence can address data deficits and mitigate health inequities in urgent and emergency care","author":"Humphries","year":"2026","journal-title":"Emerg Med J"},{"key":"ref12","volume-title":"Activity and Journeys Through Emergency and Urgent Care","year":"2017"},{"key":"ref13","volume-title":"Database Management Systems","author":"Ramakrishnan","year":"2003"},{"key":"ref14","volume-title":"Reimagining Electronic Health and Medical Records","year":"2024"},{"key":"ref15","doi-asserted-by":"publisher","first-page":"899","DOI":"10.1093\/jamia\/ocv189","article-title":"SMART on FHIR: a standards-based, interoperable apps platform for electronic health records","volume":"23","author":"Mandel","year":"2016","journal-title":"J Am Med Inform Assoc"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3447772","article-title":"Knowledge graphs","volume":"54","author":"Hogan","year":"2022","journal-title":"ACM Comput Surv"},{"key":"ref17","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1186\/s40537-023-00774-9","article-title":"Healthcare knowledge graph construction: a systematic review of the state-of-the-art, open issues, and opportunities","volume":"10","author":"Abu-Salih","year":"2023","journal-title":"J Big Data"},{"key":"ref18","doi-asserted-by":"publisher","first-page":"507","DOI":"10.1136\/jamia.2009.001560","article-title":"Mayo clinical text analysis and knowledge extraction system (cTAKES): architecture, component evaluation and applications","volume":"17","author":"Savova","year":"2010","journal-title":"J Am Med Inform Assoc"},{"key":"ref19","first-page":"279","article-title":"SNOMED-CT: the advanced terminology and coding system for eHealth","volume":"121","author":"Donnelly","year":"2006","journal-title":"Stud Health Technol Inform"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1186\/s13326-019-0198-0","article-title":"Moonstone: a novel natural language processing system for inferring social risk from clinical narratives","volume":"10","author":"Conway","year":"2019","journal-title":"J Biomed Semantics"},{"key":"ref21","first-page":"1998","volume-title":"Large Language Models Are Few-Shot Clinical Information Extractors Proceedings of EMNLP","author":"Agrawal","year":"2022"},{"key":"ref22","doi-asserted-by":"crossref","DOI":"10.64898\/2025.12.01.25341355","article-title":"CLINES: clinical LLM-based information extraction and structuring agent","author":"Yang","year":"2025"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"e281","DOI":"10.1016\/S2589-7500(24)00025-6","article-title":"Foresight - a generative pretrained transformer for modelling of patient timelines using electronic health records: a retrospective modelling study","volume":"6","author":"Kraljevic","year":"2024","journal-title":"Lancet Digit Health"},{"key":"ref24","doi-asserted-by":"publisher","first-page":"e65537","DOI":"10.2196\/65537","article-title":"Large language model-driven knowledge graph construction in sepsis care using multicenter clinical databases: development and usability study","volume":"27","author":"Yang","year":"2025","journal-title":"J Med Internet Res"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"100155","DOI":"10.1016\/j.patter.2020.100155","article-title":"KG-COVID-19: a framework to produce customised knowledge graphs for COVID-19 response","volume":"2","author":"Reese","year":"2021","journal-title":"Patterns"},{"key":"ref26","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1038\/s41597-023-01960-3","article-title":"Building a knowledge graph to enable precision medicine","volume":"10","author":"Chandak","year":"2023","journal-title":"Sci Data"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1038\/s41746-023-00879-8","article-title":"The shaky foundations of large language models and foundation models for electronic health records","volume":"6","author":"Wornow","year":"2023","journal-title":"NPJ Digit Med"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"1032","DOI":"10.1093\/jamia\/ocaf059","article-title":"Detecting emergencies in patient portal messages using large language models and knowledge graph-based retrieval-augmented generation","volume":"32","author":"Liu","year":"2025","journal-title":"J Am Med Inform Assoc"},{"key":"ref29","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1093\/jamia\/ocaf094","article-title":"SDoH-GPT: using large language models to extract social determinants of health","volume":"33","author":"Consoli","year":"2026","journal-title":"J Am Med Inform Assoc"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"103631","DOI":"10.1016\/j.jbi.2020.103631","article-title":"Annotating social determinants of health using active learning, and characterizing determinants using neural event extraction","author":"Lybarger","year":"2021","journal-title":"J Biomed Inform"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"447","DOI":"10.1126\/science.aax2342","article-title":"Dissecting racial bias in an algorithm used to manage the health of populations","volume":"366","author":"Obermeyer","year":"2019","journal-title":"Science"},{"key":"ref32","doi-asserted-by":"crossref","DOI":"10.69985\/TBQL1779","volume-title":"Fairness in Precision Medicine","author":"Ferryman","year":"2018"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"719","DOI":"10.1038\/s41551-023-01056-8","article-title":"Algorithmic fairness in artificial intelligence for medicine and healthcare","volume":"7","author":"Chen","year":"2023","journal-title":"Nat Biomed Eng"},{"key":"ref34","doi-asserted-by":"publisher","first-page":"866","DOI":"10.7326\/M18-1990","article-title":"Ensuring fairness in machine learning to advance health equity","volume":"169","author":"Rajkomar","year":"2018","journal-title":"Ann Intern Med"},{"key":"ref35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3571730","article-title":"Survey of hallucination in natural language generation","volume":"55","author":"Ji","year":"2023","journal-title":"ACM Comput Surv"},{"key":"ref36","volume-title":"Uniting the UK\u2019S Health Data: A Huge Opportunity for Society. Health Data Research UK","author":"Sudlow","year":"2024"},{"key":"ref37","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3604932","article-title":"Demystifying graph databases: analysis and taxonomy of data organization, system designs, and graph queries","volume":"56","author":"Besta","year":"2024","journal-title":"ACM Comput Surv"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"d3693","DOI":"10.1136\/bmj.d3693","article-title":"Why system inertia makes health reform so difficult","volume":"342","author":"Coiera","year":"2011","journal-title":"BMJ"},{"key":"ref39","volume-title":"Emergency Department Care 2023\u201324: Appendix Information","year":"2024"},{"key":"ref40","article-title":"Better, broader, safer: using health data for research and analysis. Department of Health and Social Care","author":"Goldacre","year":"2022"}],"container-title":["Frontiers in Public Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1901339\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T06:03:35Z","timestamp":1785305015000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpubh.2026.1901339\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,29]]},"references-count":40,"alternative-id":["10.3389\/fpubh.2026.1901339"],"URL":"https:\/\/doi.org\/10.3389\/fpubh.2026.1901339","relation":{},"ISSN":["2296-2565"],"issn-type":[{"value":"2296-2565","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,29]]},"article-number":"1901339"}}