{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T17:53:25Z","timestamp":1791309205458,"version":"4.3.1"},"reference-count":29,"publisher":"BMJ","funder":[{"name":"Washington Traffic Safety Commission","award":["2024-Sub-grants-4983"],"award-info":[{"award-number":["2024-Sub-grants-4983"]}]},{"DOI":"10.13039\/100000030","name":"Centers for Disease Control and Prevention","doi-asserted-by":"publisher","award":["1 R49CE003565-01"],"award-info":[{"award-number":["1 R49CE003565-01"]}],"id":[{"id":"10.13039\/100000030","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["Inj Prev"],"accepted":{"date-parts":[[2026,9,10]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Micromobility-related injuries are increasing, yet injury surveillance systems lack comprehensive data on risk factors like helmet use. Unstructured clinical notes in electronic health records contain this valuable information, which is crucial for injury prevention efforts.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We conducted a retrospective method comparison study comparing rule-based systems and machine learning models (Term Frequency-Inverse Document Frequency+eXtreme Gradient Boosting) for extracting helmet use information from emergency department triage notes. We analysed 1206 notes from micromobility-related injuries collected between 2018 and 2023 from three institutions: University of Washington Medicine facilities, Washington State Department of Health\u2019s Rapid Health Information NetwOrk system and the US Consumer Product Safety Commission\u2019s National Electronic Injury Surveillance System. Performance was evaluated using F1 score, precision and recall, with both internal and external validation (across three institutions) to assess generalisability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The rule-based system outperformed machine learning approaches, achieving F1 scores of 0.96 and 0.95 in two datasets and 0.86 in a third dataset, compared with machine learning models\u2019 F1 scores of approximately 0.80. The rule-based approach demonstrated higher precision and recall while requiring fewer computational resources. External validation confirmed the robustness of the rule-based system with minimal performance degradation across all three institutions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>Despite the current emphasis on machine learning approaches for clinical text analysis, our findings indicate that rule-based systems remain effective for helmet use identification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>For extraction tasks such as helmet use identification from triage notes, rule-based systems offer superior performance, reliability and efficiency compared with more complex machine learning models. These findings have implications for injury surveillance and prevention programme development.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/ip-2025-045761","type":"journal-article","created":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T17:20:17Z","timestamp":1791307217000},"page":"ip-2025-045761","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Towards automatic analysis of clinical notes for injuries of bicycles and other micromobility devices using natural language processing"],"prefix":"10.1136","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1215-8043","authenticated-orcid":false,"given":"Weipeng","family":"Zhou","sequence":"first","affiliation":[{"name":"Department of Biomedical Informatics and Data Science, Yale School of Medicine, New Haven, Connecticut, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kathryn G","family":"Burford","sequence":"additional","affiliation":[{"name":"Department of Epidemiology, Columbia University Mailman School of Public Health, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5076-3522","authenticated-orcid":false,"given":"Nicole G","family":"Itzkowitz","sequence":"additional","affiliation":[{"name":"Department of Epidemiology, Columbia University Mailman School of Public Health, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyao","family":"deGrauw","sequence":"additional","affiliation":[{"name":"University of Washington, Seattle, Washington, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kushang V","family":"Patel","sequence":"additional","affiliation":[{"name":"University of Washington, Seattle, Washington, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrew","family":"Rundle","sequence":"additional","affiliation":[{"name":"Department of Epidemiology, Columbia University Mailman School of Public Health, New York, New York, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stephen J","family":"Mooney","sequence":"additional","affiliation":[{"name":"University of Washington, Seattle, Washington, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,10,6]]},"reference":[{"key":"2026100610201182000_ip-2025-045761v1.1","unstructured":"National Highway Traffic Safety Administration, 'bicyclists and other cyclists traffic safety facts 2022 data (report no. dot hs 813591)'. U.S. Department of Transportation; 2023. Available: https:\/\/crashstats.nhtsa.dot.gov\/Api\/Public\/Publication\/813591#:~:text=In%202022%20there%20were%20an,people%20injured%20in%20traffic%20crashes.&text=In%202022%20the%20pedalcyclist%20fatality,higher%20for%20males%20than%20females [Accessed 11 Mar 2025]."},{"key":"2026100610201182000_ip-2025-045761v1.2","doi-asserted-by":"crossref","first-page":"1365","DOI":"10.2105\/AJPH.2024.307820","article-title":"The Burden of Injuries Associated With E-Bikes, Powered Scooters, Hoverboards, and Bicycles in the United States: 2019\u20122022","volume":"114","author":"Burford","year":"2024","journal-title":"Am J Public Health"},{"key":"2026100610201182000_ip-2025-045761v1.3","doi-asserted-by":"publisher","DOI":"10.1136\/injuryprev-2019-043418"},{"key":"2026100610201182000_ip-2025-045761v1.4","doi-asserted-by":"crossref","first-page":"2734","DOI":"10.1002\/lary.31213","article-title":"Trends in Head and Neck Injuries Related to Electric Versus Pedal Bicycle Use in the United States","volume":"134","author":"Williams","year":"2024","journal-title":"Laryngoscope"},{"key":"2026100610201182000_ip-2025-045761v1.5","doi-asserted-by":"publisher","DOI":"10.1080\/15389588.2024.2364358"},{"key":"2026100610201182000_ip-2025-045761v1.6","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2024.25981","article-title":"Use of Generative AI to Identify Helmet Status Among Patients With Micromobility-Related Injuries From Unstructured Clinical Notes","volume":"7","author":"Burford","year":"2024","journal-title":"JAMA Netw Open"},{"key":"2026100610201182000_ip-2025-045761v1.7","doi-asserted-by":"crossref","DOI":"10.2196\/49359","article-title":"Identifying Rare Circumstances Preceding Female Firearm Suicides: Validating A Large Language Model Approach","volume":"10","author":"Zhou","year":"2023","journal-title":"JMIR Ment Health"},{"key":"2026100610201182000_ip-2025-045761v1.8","doi-asserted-by":"crossref","first-page":"246","DOI":"10.1080\/10903127.2022.2072984","article-title":"Using Natural Language Processing to Examine Social Determinants of Health in Prehospital Pediatric Encounters and Associations with EMS Transport Decisions","volume":"27","author":"Lowery","year":"2023","journal-title":"Prehosp Emerg Care"},{"key":"2026100610201182000_ip-2025-045761v1.9","doi-asserted-by":"crossref","first-page":"687","DOI":"10.1080\/10903127.2022.2074180","article-title":"Identification of Factors Associated with Return of Spontaneous Circulation after Pediatric Out-of-Hospital Cardiac Arrest Using Natural Language Processing","volume":"27","author":"Harris","year":"2023","journal-title":"Prehosp Emerg Care"},{"key":"2026100610201182000_ip-2025-045761v1.10","doi-asserted-by":"crossref","first-page":"878","DOI":"10.5811\/WESTJEM.59070","article-title":"Social Determinants of Health in EMS Records: A Mixed-methods Analysis Using Natural Language Processing and Qualitative Content Analysis","volume":"24","author":"Burnett","year":"2023","journal-title":"West J Emerg Med"},{"key":"2026100610201182000_ip-2025-045761v1.11","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1093\/jamia\/ocad190","article-title":"Improving model transferability for clinical note section classification models using continued pretraining","volume":"31","author":"Zhou","year":"2023","journal-title":"J Am Med Inform Assoc"},{"key":"2026100610201182000_ip-2025-045761v1.12","unstructured":"Hedegaard H , Johnson RL , Garnett MF , et al . The International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10-CM) External Cause-of-injury Framework for Categorising Mechanism and Intent of Injury. Natl Health Stat Report 2019;1\u201322."},{"key":"2026100610201182000_ip-2025-045761v1.13","unstructured":"I. of Medicine, B. on H. C. Services, B. on H. S. Policy, and C. on H. R. and the P. of H. I . Beyond the hipaa privacy rule: Enhancing privacy, improving health through research. National Academies Press, 2009."},{"key":"2026100610201182000_ip-2025-045761v1.14","unstructured":"McKinney W . Python for data analysis: Data wrangling with pandas, numpy, and ipython. O\u2019Reilly Media, Inc, 2012."},{"key":"2026100610201182000_ip-2025-045761v1.15","doi-asserted-by":"crossref","unstructured":"Zhao G , Liu Y , Zhang W , et al . TFIDF based feature words extraction and topic modeling for short text. ICMSS 2018; Wuhan China, January 13, 2018:188\u201391. doi:10.1145\/3180374.3181354 Available: https:\/\/dl.acm.org\/doi\/proceedings\/10.1145\/3180374","DOI":"10.1145\/3180374.3181354"},{"key":"2026100610201182000_ip-2025-045761v1.16","doi-asserted-by":"crossref","first-page":"278","DOI":"10.1016\/j.amepre.2023.01.030","article-title":"Characterising Female Firearm Suicide Circumstances: A Natural Language Processing and Machine Learning Approach","volume":"65","author":"Goldstein","year":"2023","journal-title":"Am J Prev Med"},{"key":"2026100610201182000_ip-2025-045761v1.17","doi-asserted-by":"crossref","first-page":"1018","DOI":"10.1016\/j.jksuci.2018.06.011","article-title":"An efficient Map Reduce-Based Hybrid NBC-TFIDF algorithm to mine the public sentiment on diabetes mellitus \u2013 A big data approach","volume":"33","author":"Ramsingh","year":"2021","journal-title":"Journal of King Saud University - Computer and Information Sciences"},{"key":"2026100610201182000_ip-2025-045761v1.18","doi-asserted-by":"crossref","DOI":"10.3390\/ijerph17186513","article-title":"XGBoost-Based Framework for Smoking-Induced Noncommunicable Disease Prediction","volume":"17","author":"Davagdorj","year":"2020","journal-title":"Int J Environ Res Public Health"},{"key":"2026100610201182000_ip-2025-045761v1.19","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2020-039676","article-title":"Comparison of ARIMA model and XGBoost model for prediction of human brucellosis in mainland China: a time-series study","volume":"10","author":"Alim","year":"2020","journal-title":"BMJ Open"},{"key":"2026100610201182000_ip-2025-045761v1.20","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pgph.0000495","article-title":"Accuracy comparison of ARIMA and XGBoost forecasting models in predicting the incidence of COVID-19 in Bangladesh","volume":"2","author":"Rahman","year":"2022","journal-title":"PLOS Glob Public Health"},{"key":"2026100610201182000_ip-2025-045761v1.21","unstructured":"Scikit-learn: a set of python modules for machine learning and data mining. C, python. MacOS, Microsoft\u202f:: Windows, POSIX, Unix. Available: https:\/\/scikit-learn.org [Accessed 3 Nov 2024]."},{"key":"2026100610201182000_ip-2025-045761v1.22","doi-asserted-by":"crossref","unstructured":"Chen T , Guestrin C . XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2016;785\u201394. doi:10.1145\/2939672.2939785","DOI":"10.1145\/2939672.2939785"},{"key":"2026100610201182000_ip-2025-045761v1.23","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1017\/S026988890000624X","article-title":"Principles and practice in verifying rule-based systems","volume":"7","author":"Preece","year":"1992","journal-title":"Knowl Eng Rev"},{"key":"2026100610201182000_ip-2025-045761v1.24","doi-asserted-by":"crossref","first-page":"497","DOI":"10.1016\/0167-9236(94)90021-3","article-title":"Integrating artificial neural networks with rule-based expert systems","volume":"11","author":"Yoon","year":"1994","journal-title":"Decis Support Syst"},{"key":"2026100610201182000_ip-2025-045761v1.25","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2023.5870","article-title":"Classifying Firearm Injury Intent in Electronic Hospital Records Using Natural Language Processing","volume":"6","author":"MacPhaul","year":"2023","journal-title":"JAMA Netw Open"},{"key":"2026100610201182000_ip-2025-045761v1.26","unstructured":"Hello gpt-4o | openai. Available: https:\/\/openai.com\/index\/hello-gpt-4o\/ [Accessed 15 Feb 2025]."},{"key":"2026100610201182000_ip-2025-045761v1.27","unstructured":"'Introducing meta llama 3: the most capable openly available llm to date,' meta ai. Available: https:\/\/ai.meta.com\/blog\/meta-llama-3\/ [Accessed 13 Jul 2024]."},{"key":"2026100610201182000_ip-2025-045761v1.28","doi-asserted-by":"publisher","DOI":"10.1017\/S1351324916000334"},{"key":"2026100610201182000_ip-2025-045761v1.29","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2008.239"}],"container-title":["Injury Prevention"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/ip-2025-045761","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,6]],"date-time":"2026-10-06T17:20:21Z","timestamp":1791307221000},"score":1,"resource":{"primary":{"URL":"https:\/\/injuryprevention.bmj.com\/lookup\/doi\/10.1136\/ip-2025-045761"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,6]]},"references-count":29,"alternative-id":["10.1136\/ip-2025-045761"],"URL":"https:\/\/doi.org\/10.1136\/ip-2025-045761","relation":{},"ISSN":["1353-8047","1475-5785"],"issn-type":[{"value":"1353-8047","type":"print"},{"value":"1475-5785","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,6]]}}}