{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T22:13:58Z","timestamp":1782512038355,"version":"3.54.5"},"reference-count":102,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"10","license":[{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE J. Biomed. Health Inform."],"published-print":{"date-parts":[[2025,10]]},"DOI":"10.1109\/jbhi.2025.3574254","type":"journal-article","created":{"date-parts":[[2025,5,27]],"date-time":"2025-05-27T13:22:09Z","timestamp":1748352129000},"page":"7657-7667","source":"Crossref","is-referenced-by-count":7,"title":["Comparing Text-Based Clinical Risk Prediction in Critical Care: A Note-Specific Hierarchical Network and Large Language Models"],"prefix":"10.1109","volume":"29","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7945-4165","authenticated-orcid":false,"given":"Jinghui","family":"Liu","sequence":"first","affiliation":[{"name":"Australian e-Health Research Centre, CSIRO, Brisbane, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6215-6954","authenticated-orcid":false,"given":"Anthony","family":"Nguyen","sequence":"additional","affiliation":[{"name":"Australian e-Health Research Centre, CSIRO, Brisbane, QLD, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9256-1256","authenticated-orcid":false,"given":"Daniel","family":"Capurro","sequence":"additional","affiliation":[{"name":"School of Computing and Information Systems and Centre for Digital Transformation of Health, University of Melbourne, Melbourne, VIC, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8661-1544","authenticated-orcid":false,"given":"Karin","family":"Verspoor","sequence":"additional","affiliation":[{"name":"School of Computing Technologies, RMIT University, Melbourne, VIC, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"crossref","DOI":"10.1007\/978-0-387-77244-8","volume-title":"Clinical Prediction Models","author":"Steyerberg","year":"2009"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1056\/nejm198405173102001"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00832-9"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-021-01469-6"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1377\/hlthaff.2014.0041"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/2623330.2623742"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/jbhi.2017.2767063"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0300-7"},{"key":"ref9","first-page":"446","article-title":"A multivariate timeseries modeling approach to severity of illness assessment and forecasting in ICU with sparse, heterogeneous clinical data","volume-title":"Proc. AAAI Conf. Artif. Intell. Conf. Artif. Intell.","author":"Ghassemi","year":"2015"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-018-0029-1"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.m1328"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2009.08.007"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2013.393"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2019.00066"},{"key":"ref15","first-page":"26","article-title":"Whats in a note? Unpacking predictive value in clinical note representations","volume-title":"Proc. AMIA Joint Summits Transl. Sci.","volume":"2017","author":"Boag","year":"2018"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocac058"},{"key":"ref17","first-page":"4026","article-title":"Predicting in-hospital mortality by combining clinical notes with time-series data","volume-title":"Proc. Findings Assoc. Comput. Linguistics","author":"Deznabi","year":"2021"},{"key":"ref18","article-title":"LCD benchmark: Long clinical document benchmark on mortality prediction for language models","volume":"32","author":"Yoon","year":"2024","journal-title":"J. Amer. Med. Inform. Assoc."},{"key":"ref19","first-page":"440","article-title":"Deep EHR: Chronic disease prediction using medical notes","volume-title":"Proc. 3rd Mach. Learn. Healthcare Conf.","author":"Liu","year":"2018"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocaa004"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-021-00474-9"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00989-3"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.4172\/2157-7420.s3-001"},{"key":"ref24","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Brown","year":"2020"},{"key":"ref25","article-title":"Finetuned language models are zero-shot learners","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Wei","year":"2022"},{"key":"ref26","article-title":"Training language models to follow instructions with human feedback","author":"Ouyang","year":"2022"},{"key":"ref27","doi-asserted-by":"crossref","first-page":"2114","DOI":"10.1093\/jamia\/ocae074","article-title":"Large language models for biomedicine: Foundations, opportunities, challenges, and best practices","volume":"31","author":"Sahoo","year":"2024","journal-title":"J. Amer. Med. Inform. Assoc."},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.7326\/m23-2772"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-023-02448-8"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06291-2"},{"issue":"2","key":"ref31","doi-asserted-by":"crossref","DOI":"10.1056\/AIoa2300068","article-title":"Almanac: Retrieval-augmented language models for clinical medicine","volume":"1","author":"Zakka","year":"2024","journal-title":"NEJM AI"},{"key":"ref32","article-title":"Capabilities of GPT-4 on medical challenge problems","author":"Nori","year":"2023"},{"key":"ref33","doi-asserted-by":"crossref","first-page":"319","DOI":"10.1001\/jama.2024.21700","article-title":"Testing and evaluation of health care applications of large language models: A systematic review","volume":"333","author":"Bedi","year":"2024","journal-title":"J. Amer. Med. Assoc."},{"issue":"8","key":"ref34","doi-asserted-by":"crossref","first-page":"928","DOI":"10.1001\/jamasurg.2024.1621","article-title":"Large language model capabilities in perioperative risk prediction and prognostication","volume":"159","author":"Chung","year":"2024","journal-title":"JAMA Surg."},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.8895"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-022-00616-7"},{"key":"ref37","first-page":"779","article-title":"Deep patient representation of clinical notes via multi-task learning for mortality prediction","volume-title":"Proc. AMIA Joint Summits Transl. Sci.","volume":"2019","author":"Si","year":"2019"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1186\/s13326-021-00235-3"},{"key":"ref39","doi-asserted-by":"crossref","first-page":"4190","DOI":"10.18653\/v1\/2020.acl-main.385","article-title":"Quantifying attention flow in transformers","volume-title":"Proc. 58th Annu. Meeting Assoc. Comput. Linguistics","author":"Abnar","year":"2020"},{"key":"ref40","first-page":"438","article-title":"Literature-augmented clinical outcome prediction","volume-title":"Proc. Findings Assoc. Comput. Linguistics","author":"Naik","year":"2022"},{"key":"ref41","first-page":"73","article-title":"Improving text-based early prediction by distillation from privileged time-series text","volume-title":"Proc. 20th Annu. Workshop Australas. Lang. Technol. Assoc.","author":"Liu","year":"2022"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-021-00426-3"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-022-00689-4"},{"key":"ref44","first-page":"1478","article-title":"Explainable clinical decision support from text","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process.","author":"Feng","year":"2020"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-023-43713-1"},{"key":"ref46","first-page":"2062","article-title":"Characterizing the value of information in medical notes","volume-title":"Proc. Findings Assoc. Comput. Linguistics","author":"Hsu","year":"2020"},{"key":"ref47","doi-asserted-by":"crossref","DOI":"10.1016\/j.jbi.2023.104466","article-title":"Attention-based multimodal fusion with contrast for robust clinical prediction in the face of missing modalities","volume":"145","author":"Liu","year":"2023","journal-title":"J. Biomed. Inform."},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1056\/nejmc2104626"},{"key":"ref49","first-page":"506","article-title":"Mind the performance gap: Examining dataset shift during prospective validation","volume-title":"Proc. 6th Mach. Learn. Healthcare Conf.","author":"Otles","year":"2021"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pdig.0000198"},{"key":"ref51","doi-asserted-by":"crossref","DOI":"10.1093\/jamia\/ocae085","article-title":"Local large language models for privacy-preserving accelerated review of historic echocardiogram reports","volume":"31","author":"Vaid","year":"2024","journal-title":"J. Amer. Med. Inform. Assoc."},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01091-y"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2016.35"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.3233\/shti231244"},{"key":"ref55","article-title":"Modular deep learning","author":"Pfeiffer","year":"2023","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref56","first-page":"5557","article-title":"DEMix layers: Disentangling domains for modular language modeling","volume-title":"Proc. Conf. North Amer. Chapter Assoc. Comput. Linguistics: Hum. Lang. Technol.","author":"Gururangan","year":"2022"},{"key":"ref57","article-title":"Outrageously large neural networks: The sparsely-gated mixture-of-experts layer","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Shazeer","year":"2017"},{"key":"ref58","article-title":"Mixtral of experts","author":"Jiang","year":"2024"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2022.104149"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1145\/3368555.3384469"},{"key":"ref61","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023"},{"key":"ref62","article-title":"The Llama 3 herd of models","year":"2024"},{"key":"ref63","article-title":"Mistral 7B","author":"Jiang","year":"2023"},{"key":"ref64","article-title":"Qwen2.5 technical report","author":"Team","year":"2024"},{"key":"ref65","article-title":"Qwen2 technical report","author":"Yang","year":"2024"},{"key":"ref66","article-title":"Phi-3 technical report: A highly capable language model locally on your phone","author":"Abdin","year":"2024"},{"key":"ref67","article-title":"Gemma 2: Improving open language models at a practical size","year":"2024"},{"key":"ref68","article-title":"MedAlpacaan open-source collection of medical conversational AI models and training data","author":"Han","year":"2023"},{"key":"ref69","article-title":"MEDITRON-70B: Scaling medical pretraining for large language models","author":"Chen","year":"2023"},{"key":"ref70","article-title":"UltraMedical: Building specialized generalists in biomedicine","author":"Zhang","year":"2024"},{"key":"ref71","first-page":"24824","article-title":"Chain of thought prompting elicits reasoning in large language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wei","year":"2022"},{"key":"ref72","article-title":"Large language models are zero-shot reasoners","author":"Kojima","year":"2022"},{"key":"ref73","article-title":"Self-consistency improves chain of thought reasoning in language models","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Wang","year":"2023"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2020.103540"},{"issue":"3","key":"ref75","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1093\/jamia\/ocae321","article-title":"Factors underpinning the performance of implemented artificial intelligence-based patient deterioration prediction systems: Reasons for selection and implications for hospitals and researchers","volume":"32","author":"der Vegt","year":"2025","journal-title":"J. Amer. Med. Inform. Assoc."},{"key":"ref76","first-page":"356","article-title":"How noisy social media text, how diffrnt social media sources?","volume-title":"Proc. 6th Int. Joint Conf. Natural Lang. Process.","author":"Baldwin","year":"2013"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1075\/ijcl.6.1.05kil"},{"key":"ref78","first-page":"1675","article-title":"Cost-effective selection of pretraining data: A case study of pretraining BERT on social media","volume-title":"Proc. Findings Assoc. Comput. Linguistics","author":"Dai","year":"2020"},{"key":"ref79","first-page":"4171","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","volume-title":"Proc. Conf. North Amer. Chapter Assoc. Comput. Linguistics: Hum. Lang. Technol.","author":"Devlin","year":"2019"},{"key":"ref80","article-title":"On the stability of fine-tuning BERT: Misconceptions, explanations, and strong baselines","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Mosbach","year":"2021"},{"key":"ref81","first-page":"5747","article-title":"Understanding the difficulty of training transformers","volume-title":"Proc. Conf. Empirical Methods Natural Lang. Process.","author":"Liu","year":"2020"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-018-24271-9"},{"key":"ref83","first-page":"597","article-title":"Patient representation transfer learning from clinical notes based on hierarchical attention network","volume-title":"Proc. AMIA Joint Summits Transl. Sci.","volume":"2020","author":"Si","year":"2020"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2021.103726"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6331"},{"key":"ref86","doi-asserted-by":"crossref","first-page":"72","DOI":"10.18653\/v1\/W19-1909","article-title":"Publicly available clinical BERT embeddings","volume-title":"Proc. 2nd Clin. Natural Lang. Process. Workshop","author":"Alsentzer","year":"2019"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1145\/3458754"},{"key":"ref88","article-title":"Longformer: The long-document transformer","author":"Beltagy","year":"2020"},{"key":"ref89","first-page":"17283","article-title":"Big bird: Transformers for longer sequences","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zaheer","year":"2020"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocac225"},{"key":"ref91","article-title":"LoRA: Low-rank adaptation of large language models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Hu","year":"2022"},{"key":"ref92","first-page":"10088","article-title":"QLoRA: Efficient finetuning of quantized LLMs","volume-title":"Proc. 37th Conf. Neural Inf. Process. Syst.","author":"Dettmers","year":"2023"},{"key":"ref93","first-page":"807","article-title":"Computer-assisted diagnostic coding: Effectiveness of an NLP-based approach using SNOMED CT to ICD-10 mappings","volume-title":"Proc. Annu. Symp. Proc. \/ AMIA Symp.","author":"Nguyen","year":"2018"},{"key":"ref94","first-page":"104","article-title":"Making the most out of the limited context length: Predictive power varies with clinical note type and note section","volume-title":"Proc. 61st Annu. Meeting Assoc. Comput. Linguistics","author":"Zheng","year":"2023"},{"issue":"7969","key":"ref95","first-page":"357","article-title":"Health system-scale language models are all-purpose prediction engines","volume-title":"Nature","volume":"619","author":"Jiang","year":"2023"},{"key":"ref96","article-title":"RULER: Whats the real context size of your long-context language models?","volume-title":"Proc. 1st Conf. Lang. Model.","author":"Hsieh","year":"2024"},{"key":"ref97","first-page":"49","article-title":"Attention flows are shapley value explanations","volume-title":"Proc. 59th Annu. Meeting Assoc. Comput. Linguistics 11th Int. Joint Conf. Natural Lang. Process.","author":"Ethayarajh","year":"2021"},{"key":"ref98","doi-asserted-by":"crossref","DOI":"10.1093\/jamia\/ocaf038","article-title":"Large language models are less effective at clinical prediction tasks than locally trained machine learning models","volume":"32","author":"Brown","year":"2025","journal-title":"J. Amer. Med. Inform. Assoc."},{"key":"ref99","doi-asserted-by":"crossref","first-page":"157","DOI":"10.1162\/tacl_a_00638","article-title":"Lost in the middle: How language models use long contexts","volume":"12","author":"Liu","year":"2024","journal-title":"Trans. Assoc. Comput. Linguistics"},{"key":"ref100","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-022-00730-6"},{"key":"ref101","first-page":"4765","article-title":"A unified approach to interpreting model predictions","volume-title":"Proc. Neural Inf. Process. Syst.","author":"Lundberg","year":"2017"},{"key":"ref102","article-title":"Judging LLM-as-a-judge with MT-bench and chatbot arena","author":"Zheng","year":"2023"}],"container-title":["IEEE Journal of Biomedical and Health Informatics"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6221020\/11192794\/11016178.pdf?arnumber=11016178","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,7]],"date-time":"2025-10-07T05:25:02Z","timestamp":1759814702000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11016178\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10]]},"references-count":102,"journal-issue":{"issue":"10"},"URL":"https:\/\/doi.org\/10.1109\/jbhi.2025.3574254","relation":{},"ISSN":["2168-2194","2168-2208"],"issn-type":[{"value":"2168-2194","type":"print"},{"value":"2168-2208","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10]]}}}