{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T06:11:11Z","timestamp":1783923071527,"version":"3.55.0"},"reference-count":56,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T00:00:00Z","timestamp":1783900800000},"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. Digit. Health"],"abstract":"\n Introduction<\/jats:title>\n Geriatric syndromes (GS), such as falls, dementia, delirium and malnutrition, are complex clinical conditions affecting older adults which involve multiple organ systems and have major impact on quality of life and care. GS cut across disease categories, and are poorly represented in structured electronic health records. Natural language processing (NLP) offers an opportunity to extract valuable GS-related information from unstructured clinical text, such as hospital discharge summaries. However, the lack of high-quality annotated datasets limits the effectiveness of NLP models in this domain. This study introduces a manually annotated corpus designed for GS detection, enabling more accurate identification and classification of GS.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We developed a comprehensive and detailed annotation scheme to label 12 common GS from hospital discharge summaries, incorporating key attributes such as diagnosis type, negation and event occurrence. The corpus consists of 2,040 manually annotated discharge summaries from National Health Service (NHS) Lothian hospitals in Scotland. To assess the effectiveness of NLP in extracting GS, we experimented with multiple pretrained transformer-based models, including base BERT (general-domain), BioBERT (biomedical-domain), BioClinicalBERT (clinical-domain) and BERT-cased (the cased English BERT checkpoint). The models were fine-tuned and tested for two types of tasks: named entity recognition (NER) and document-level labelling. We also considered an extra task of detecting contextual information with each GS mention (e.g., history, suspected, in-hospital). When context information is considered, two new tasks are called NER-C and DL-C, for NER and document-level labelling with context respectively.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n Our evaluation showed that, for the document-level labelling task, BERT-cased achieved the highest F1-score (0.897) and BioClinicalBERT performed best when negation was considered (F1-score: 0.888). For the NER task, BioClinicalBERT and BERT-cased achieved an F1-score of 0.883. Frailty (F1\n \n \n =<\/mml:mo>\n <\/mml:math>\n <\/jats:inline-formula>\n 1.0), Falls (F1\n \n \n =<\/mml:mo>\n <\/mml:math>\n <\/jats:inline-formula>\n 0.973) and Delirium (F1\n \n \n =<\/mml:mo>\n <\/mml:math>\n <\/jats:inline-formula>\n 0.946) are the GS entities with the best performing results. For NER-C, BERT-cased achieved the best F1 of 0.692 and BioBERT performed the worst (F1\n \n \n =<\/mml:mo>\n <\/mml:math>\n <\/jats:inline-formula>\n 0.658). In NER-C, the best results were achieved for context-aware falls and frailty labels, particularly when the syndrome was implied rather than explicitly stated. Document-level aggregation helped reduce inconsistencies, but the NER experiments used a flattened CoNLL-compatible representation of the original annotations, in which discontinuous mentions were converted into shortest covering spans and overlapping mentions were merged. Therefore, the reported NER results should be interpreted as baseline performance on a simplified representation of these structures, while low-frequency GS categories and sparse contextual labels also negatively affected model accuracy.\n <\/jats:p>\n <\/jats:sec>\n \n Discussion<\/jats:title>\n This study demonstrates the effectiveness of NLP for extracting geriatric syndromes from unstructured clinical text and introduces a manually annotated corpus with detailed guidelines to support this task. The results also show that model performance is strongly shaped by dataset characteristics. More frequent and lexically clearer syndromes, such as frailty, falls and delirium, achieved the strongest results, whereas rarer categories and low-frequency attribute combinations, such as suspected, referral and some negated or context-specific labels, were harder to learn and yielded lower and less stable scores. Likewise, fine-grained annotation was more challenging than coarse-grained annotation because it increases label sparsity and requires the model to distinguish subtle contextual differences, such as current vs. historical mentions, implicit mentions and in-hospital onset. Entity-level extraction was further affected by discontinuous and overlapping mentions, which are common in clinical narratives and make boundary detection harder, whereas document-level aggregation reduced the impact of these local errors and therefore produced higher scores. These findings underline that data distribution, annotation complexity and mention structure directly influence model performance, and should be central considerations in future work on geriatric syndrome extraction.<\/jats:p>\n <\/jats:sec>","DOI":"10.3389\/fdgth.2026.1591050","type":"journal-article","created":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T05:35:14Z","timestamp":1783920914000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Geriatric syndromes extraction from discharge summaries: a new dataset, annotation scheme and initial findings"],"prefix":"10.3389","volume":"8","author":[{"given":"Imane","family":"Guellil","sequence":"first","affiliation":[{"name":"Advanced Care Research Centre - Usher Institute, University of Edinburgh","place":["Edinburgh, United Kingdom"]},{"name":"Department of Cancer and Genomic Sciences, University of Birmingham","place":["Birmingham, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Salom\u00e9","family":"Andres","sequence":"additional","affiliation":[{"name":"Advanced Care Research Centre - Usher Institute, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Atul","family":"Anand","sequence":"additional","affiliation":[{"name":"Advanced Care Research Centre - Usher Institute, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bruce","family":"Guthrie","sequence":"additional","affiliation":[{"name":"Advanced Care Research Centre - Usher Institute, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fahrurrozi","family":"Rahman","sequence":"additional","affiliation":[{"name":"Advanced Care Research Centre - Usher Institute, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Abul","family":"Hasan","sequence":"additional","affiliation":[{"name":"Institute of Health Informatics, University College London (UCL)","place":["London, United Kingdom"]},{"name":"Nuffield Department of Primary Care Health Sciences, Oxford University","place":["Oxford, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huayu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Advanced Care Research Centre - Usher Institute, University of Edinburgh","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Honghan","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Health & Wellbeing, University of Glasgow","place":["Glasgow, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Beatrice","family":"Alex","sequence":"additional","affiliation":[{"name":"Advanced Care Research Centre - Usher Institute, University of Edinburgh","place":["Edinburgh, United Kingdom"]},{"name":"School of Mathematical & Computer Sciences, Heriot-Watt University","place":["Edinburgh, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,13]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"780","DOI":"10.1111\/j.1532-5415.2007.01156.x","article-title":"Geriatric syndromes: clinical, research, and policy implications of a core geriatric concept: (see editorial comments by Dr. William Hazzard on pp 794\u2013796)","volume":"55","author":"Inouye","year":"2007","journal-title":"J Am Geriatr Soc"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1499","DOI":"10.1111\/jgs.15411","article-title":"The value of unstructured electronic health record data in geriatric syndrome case identification","volume":"66","author":"Kharrazi","year":"2018","journal-title":"J Am Geriatr Soc"},{"key":"B3","doi-asserted-by":"publisher","first-page":"103851","DOI":"10.1016\/j.jbi.2021.103851","article-title":"Adaptation of an NLP system to a new healthcare environment to identify social determinants of health","volume":"120","author":"Reeves","year":"2021","journal-title":"J Biomed Inform"},{"key":"B4","doi-asserted-by":"publisher","first-page":"e052735","DOI":"10.1136\/bmjopen-2021-052735","article-title":"Variation in coded frailty syndromes in secondary care administrative data: an international retrospective exploratory study","volume":"12","author":"Soong","year":"2022","journal-title":"BMJ Open"},{"key":"B5","first-page":"34","article-title":"Coding geriatric syndromes: how good are we?","volume":"10","author":"Ugboma","year":"2008","journal-title":"CME J Geriatr Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"103833","DOI":"10.1016\/j.jbi.2021.103833","article-title":"Augmenting aer2vec: enriching distributed representations of adverse event report data with orthographic and lexical information","volume":"119","author":"Ding","year":"2021","journal-title":"J Biomed Inform"},{"key":"B7","doi-asserted-by":"publisher","first-page":"2663","DOI":"10.3390\/app11062663","article-title":"Automatic extraction of adverse drug reactions from summary of product characteristics","volume":"11","author":"Shen","year":"2021","journal-title":"Appl Sci"},{"key":"B8","doi-asserted-by":"publisher","first-page":"e18471","DOI":"10.2196\/18471","article-title":"Defining patient-oriented natural language processing: a new paradigm for research and development to facilitate adoption and use by medical experts","volume":"9","author":"Sarker","year":"2021","journal-title":"JMIR Med Inform"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12877-017-0645-7","article-title":"Comparing clinician descriptions of frailty and geriatric syndromes using electronic health records: a retrospective cohort study","volume":"17","author":"Anzaldi","year":"2017","journal-title":"BMC Geriatr"},{"key":"B10","doi-asserted-by":"publisher","first-page":"e13039","DOI":"10.2196\/13039","article-title":"Extraction of geriatric syndromes from electronic health record clinical notes: assessment of statistical natural language processing methods","volume":"7","author":"Chen","year":"","journal-title":"JMIR Med Inform"},{"key":"B11","first-page":"16","article-title":"Study of the association of geriatric syndromes with functional dependence in the elderly","volume":"7","author":"Neeta","year":"2018","journal-title":"Int J Med Health Sci"},{"key":"B12","author":"Stanovsky","year":""},{"key":"B13","author":"Dai","year":""},{"key":"B14","doi-asserted-by":"publisher","first-page":"e48995","DOI":"10.2196\/48995","article-title":"Bert-based neural network for inpatient fall detection from electronic medical records: retrospective cohort study","volume":"12","author":"Cheligeer","year":"2024","journal-title":"JMIR Med Inform"},{"key":"B15","author":"dos Santos","year":""},{"key":"B16","doi-asserted-by":"publisher","first-page":"105401","DOI":"10.1016\/j.ebiom.2024.105401","article-title":"Enhancing early detection of cognitive decline in the elderly: a comparative study utilizing large language models in clinical notes","volume":"109","author":"Du","year":"2024","journal-title":"eBioMedicine"},{"key":"B17","doi-asserted-by":"publisher","first-page":"524","DOI":"10.1093\/gerona\/glaa275","article-title":"Ascertainment of delirium status using natural language processing from electronic health records","volume":"77","author":"Fu","year":"2020","journal-title":"J Gerontol Ser A"},{"key":"B18","doi-asserted-by":"publisher","first-page":"104736","DOI":"10.1016\/j.ijmedinf.2022.104736","article-title":"A hybrid model to identify fall occurrence from electronic health records","volume":"162","author":"Fu","year":"2022","journal-title":"Int J Med Inform"},{"key":"B19","doi-asserted-by":"publisher","first-page":"e33834","DOI":"10.2196\/33834","article-title":"Identifying patients with delirium based on unstructured clinical notes: observational study","volume":"6","author":"Ge","year":"2022","journal-title":"JMIR Form Res"},{"key":"B20","doi-asserted-by":"publisher","first-page":"109","DOI":"10.1093\/jamia\/ocab248","article-title":"Development and validation of a prediction model for actionable aspects of frailty in the text of clinicians\u2019 encounter notes","volume":"29","author":"Martin","year":"2021","journal-title":"J Am Med Inform Assoc"},{"key":"B21","doi-asserted-by":"publisher","first-page":"e16970","DOI":"10.2196\/16970","article-title":"Predicting inpatient falls using natural language processing of nursing records obtained from Japanese electronic medical records: case-control study","volume":"8","author":"Nakatani","year":"","journal-title":"JMIR Med Inform"},{"key":"B22","doi-asserted-by":"publisher","first-page":"167","DOI":"10.3390\/bdcc7040167","article-title":"Identifying probable dementia in undiagnosed black and white Americans using machine learning in veterans health administration electronic health records","volume":"7","author":"Shao","year":"2023","journal-title":"Big Data Cogn Comput"},{"key":"B23","doi-asserted-by":"publisher","first-page":"e38161","DOI":"10.2196\/38161","article-title":"Boosting delirium identification accuracy with sentiment-based natural language processing: mixed methods study","volume":"10","author":"Wang","year":"2022","journal-title":"JMIR Med Inform"},{"key":"B24","doi-asserted-by":"publisher","first-page":"afad046","DOI":"10.1093\/ageing\/afad046","article-title":"Predicting future falls in older people using natural language processing of general practitioners\u2019 clinical notes","volume":"52","author":"Dormosh","year":"2023","journal-title":"Age Ageing"},{"key":"B25","author":"Mishra","year":""},{"key":"B26","doi-asserted-by":"publisher","first-page":"e13039","DOI":"10.2196\/13039","article-title":"Extraction of geriatric syndromes from electronic health record clinical notes: assessment of statistical natural language processing methods","volume":"7","author":"Chen","year":"","journal-title":"JMIR Med Inform"},{"key":"B27","doi-asserted-by":"publisher","first-page":"4406","DOI":"10.3390\/app11104406","article-title":"Does the integration of pre-coded information with narratives improve in-hospital falls\u2019 surveillance?","volume":"11","author":"Lorenzoni","year":"2021","journal-title":"Appl Sci"},{"key":"B28","doi-asserted-by":"publisher","first-page":"152","DOI":"10.1111\/jnu.13038","article-title":"Applying natural language processing to understand symptoms among older adult home healthcare patients with urinary incontinence","volume":"57","author":"Scharp","year":"2025","journal-title":"J Nurs Scholarsh"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.1002\/gps.5659","article-title":"Weight loss in Alzheimer\u2019s disease, vascular dementia and dementia with Lewy bodies: Impact on mortality and hospitalization by dementia subtype","volume":"37","author":"Soysal","year":"2022","journal-title":"Int J Geriatr Psychiatry"},{"key":"B30","doi-asserted-by":"publisher","first-page":"e17819","DOI":"10.2196\/17819","article-title":"Predicting onset of dementia using clinical notes and machine learning: case-control study","volume":"8","author":"Hane","year":"2020","journal-title":"JMIR Med Inform"},{"key":"B31","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1007\/s41666-023-00125-6","article-title":"Can patients with dementia be identified in primary care electronic medical records using natural language processing?","volume":"7","author":"Maclagan","year":"2023","journal-title":"J Healthc Inform Res"},{"key":"B32","doi-asserted-by":"publisher","first-page":"2267","DOI":"10.3233\/THC-230229","article-title":"Malnutrition and its contributing factors for older people living in residential aged care facilities: insights from natural language processing of aged care records","volume":"31","author":"Alkhalaf","year":"2023","journal-title":"Technol Health Care"},{"key":"B33","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1186\/s12911-022-01864-z","article-title":"Development of a machine learning model to predict mild cognitive impairment using natural language processing in the absence of screening","volume":"22","author":"Penfold","year":"2022","journal-title":"BMC Med Inform Decis Mak"},{"key":"B34","doi-asserted-by":"publisher","first-page":"160035","DOI":"10.1038\/sdata.2016.35","article-title":"MIMIC-III, a freely accessible critical care database","volume":"3","author":"Johnson","year":"2016","journal-title":"Sci Data"},{"key":"B35","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41597-022-01899-x","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci Data"},{"key":"B36","doi-asserted-by":"publisher","first-page":"552","DOI":"10.1136\/amiajnl-2011-000203","article-title":"2010 i2b2\/VA challenge on concepts, assertions, and relations in clinical text","volume":"18","author":"Uzuner","year":"2011","journal-title":"J Am Med Inform Assoc"},{"key":"B37","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1093\/jamia\/ocz166","article-title":"2018 n2c2 shared task on adverse drug events and medication extraction in electronic health records","volume":"27","author":"Henry","year":"2020","journal-title":"J Am Med Inform Assoc"},{"key":"B38","doi-asserted-by":"publisher","first-page":"e16970","DOI":"10.2196\/16970","article-title":"Predicting inpatient falls using natural language processing of nursing records obtained from Japanese electronic medical records: case-control study","volume":"8","author":"Nakatani","year":"","journal-title":"JMIR Med Inform"},{"key":"B39","doi-asserted-by":"publisher","first-page":"1265","DOI":"10.1162\/tacl_a_00602","article-title":"T\n 2\n \n \n -NER: a two-stage span-based framework for unified named entity recognition with templates","volume":"11","author":"Huang","year":"2023","journal-title":"Trans Assoc Comput Linguist"},{"key":"B40","author":"Stenetorp","year":""},{"key":"B41","doi-asserted-by":"publisher","first-page":"e11990","DOI":"10.2196\/11990","article-title":"Detecting hypoglycemia incidents reported in patients\u2019 secure messages: using cost-sensitive learning and oversampling to reduce data imbalance","volume":"21","author":"Chen","year":"","journal-title":"J Med Internet Res"},{"key":"B42","author":"Alex","year":""},{"key":"B43","author":"Akbik","year":""},{"key":"B44","author":"Pennington","year":""},{"key":"B45","author":"Devlin","year":""},{"key":"B46","doi-asserted-by":"publisher","first-page":"1234","DOI":"10.1093\/bioinformatics\/btz682","article-title":"BioBERT: a pre-trained biomedical language representation model for biomedical text mining","volume":"36","author":"Lee","year":"2020","journal-title":"Bioinformatics"},{"key":"B47","author":"Alsentzer","year":""},{"key":"B48","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","author":"Hochreiter","year":"1997","journal-title":"Neural Comput"},{"key":"B49","author":"Lafferty","year":""},{"key":"B50","doi-asserted-by":"crossref","DOI":"10.1201\/9781420049176","volume-title":"Recurrent Neural Networks: Design and Applications","author":"Medsker","year":"1999"},{"key":"B51","doi-asserted-by":"publisher","first-page":"101","DOI":"10.1214\/ss\/1009213286","article-title":"Interval estimation for a binomial proportion","volume":"16","author":"Brown","year":"2001","journal-title":"Stat Sci"},{"key":"B52","doi-asserted-by":"publisher","first-page":"e0303653","DOI":"10.1371\/journal.pone.0303653","article-title":"Identifying stigmatizing language in clinical documentation: a scoping review of emerging literature","volume":"19","author":"Barcelona","year":"2024","journal-title":"PLoS ONE"},{"key":"B53","doi-asserted-by":"publisher","first-page":"e50428","DOI":"10.2196\/50428","article-title":"Examining linguistic differences in electronic health records for diverse patients with diabetes: natural language processing analysis","volume":"12","author":"Bilotta","year":"2024","journal-title":"JMIR Med Inform"},{"key":"B54","doi-asserted-by":"publisher","first-page":"e2144967","DOI":"10.1001\/jamanetworkopen.2021.44967","article-title":"Examination of stigmatizing language in the electronic health record","volume":"5","author":"Himmelstein","year":"2022","journal-title":"JAMA Netw Open"},{"key":"B55","doi-asserted-by":"publisher","first-page":"203","DOI":"10.1377\/hlthaff.2021.01423","article-title":"Negative patient descriptors: documenting racial bias in the electronic health record","volume":"41","author":"Sun","year":"2022","journal-title":"Health Aff"},{"key":"B56","author":"Dai","year":""}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1591050\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T05:35:16Z","timestamp":1783920916000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1591050\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,13]]},"references-count":56,"alternative-id":["10.3389\/fdgth.2026.1591050"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1591050","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,13]]},"article-number":"1591050"}}