{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T19:00:13Z","timestamp":1790190013573,"version":"4.1.0"},"reference-count":54,"publisher":"Oxford University Press (OUP)","issue":"1","license":[{"start":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T00:00:00Z","timestamp":1785888000000},"content-version":"vor","delay-in-days":216,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,1,7]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>This study proposes a multimodal framework for retrospective short-term mortality prediction in intensive care unit patients by integrating structured admission data with semantic information extracted from discharge summaries. The objective is to improve patient characterization by combining complementary clinical data that are often analyzed separately.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Structured inputs, including demographics, administrative records, laboratory results, and clinical severity scores, were merged with semantic representations of clinical notes produced by a BERT-based encoder. The resulting embeddings were reduced in dimensionality and clustered with K-Means to obtain categorical semantic features reflecting thematic similarity among notes. These semantic cluster identifiers were then concatenated with tabular variables to create a unified multimodal representation, which was used to train five classical machine learning classifiers.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The integration of semantic features consistently improved predictive performance compared with models trained only on structured data. Among the evaluated algorithms, Random Forest achieved the best discrimination, with an area under the receiver operating characteristic curve of 0.940 in the multimodal setting.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Combining structured clinical variables with unstructured admission notes enhances short-term mortality prediction and supports the development of interpretable multimodal decision-support tools for intensive care settings.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioadv\/vbag214","type":"journal-article","created":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T12:03:53Z","timestamp":1785585833000},"source":"Crossref","is-referenced-by-count":0,"title":["ICU mortality prediction via multimodal learning on MIMIC-IV clinical data"],"prefix":"10.1093","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2660-1432","authenticated-orcid":false,"given":"Alberto","family":"Falcone","sequence":"first","affiliation":[{"name":"Institute for High Performance Computing and Networking, Italian National Research Council (CNR-ICAR) , Rende 87036,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3025-7689","authenticated-orcid":false,"given":"Agostino","family":"Forestiero","sequence":"additional","affiliation":[{"name":"Institute for High Performance Computing and Networking, Italian National Research Council (CNR-ICAR) , Rende 87036,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8313-2492","authenticated-orcid":false,"given":"Rocco","family":"Malivindi","sequence":"additional","affiliation":[{"name":"Department of Pharmacy, Health and Nutritional Sciences, University of Calabria , Rende 87036,","place":["Italy"]},{"name":"Clinical Laboratory Unit, Azienda Ospedaliera di Cosenza \u201cAnnunziata\u201d , Cosenza 87100,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5558-9033","authenticated-orcid":false,"given":"Marzia","family":"Settino","sequence":"additional","affiliation":[{"name":"Institute for High Performance Computing and Networking, Italian National Research Council (CNR-ICAR) , Rende 87036,","place":["Italy"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,8,5]]},"reference":[{"key":"2026092314110371000_vbag214-B1","doi-asserted-by":"crossref","first-page":"134","DOI":"10.24297\/jam.v21i.9279","article-title":"The use of one sample t-test in the real data","volume":"21","author":"Al-Kassab","year":"2022","journal-title":"JAM"},{"key":"2026092314110371000_vbag214-B2","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s10994-025-06867-1","article-title":"LLM-based feature generation from text for interpretable machine learning","volume":"114","author":"Balek","year":"2025","journal-title":"Mach Learn"},{"key":"2026092314110371000_vbag214-B3","doi-asserted-by":"crossref","first-page":"26","DOI":"10.1038\/s43856-024-00717-2","article-title":"Current applications and challenges in large language models for patient care: a systematic review","volume":"5","author":"Busch","year":"2025","journal-title":"Commun Med (Lond)"},{"key":"2026092314110371000_vbag214-B4","doi-asserted-by":"crossref","first-page":"101959","DOI":"10.1016\/j.jocs.2023.101959","article-title":"Online evaluation of the Kolmogorov\u2013Smirnov test on arbitrarily large samples","volume":"67","author":"Cardoso","year":"2023","journal-title":"J Comput Sci"},{"key":"2026092314110371000_vbag214-B5","doi-asserted-by":"crossref","first-page":"373","DOI":"10.1016\/0021-9681(87)90171-8","article-title":"A new method of classifying prognostic comorbidity in longitudinal studies: development and validation","volume":"40","author":"Charlson","year":"1987","journal-title":"J Chronic Dis"},{"key":"2026092314110371000_vbag214-B6","doi-asserted-by":"crossref","first-page":"1245","DOI":"10.1016\/0895-4356(94)90129-5","article-title":"Validation of a combined comorbidity index","volume":"47","author":"Charlson","year":"1994","journal-title":"J Clin Epidemiol"},{"key":"2026092314110371000_vbag214-B7","doi-asserted-by":"crossref","first-page":"5503","DOI":"10.3390\/jcm13185503","article-title":"Predicting ICU readmission from electronic health records via BERTopic with long short term memory network approach","volume":"13","author":"Chiu","year":"2024","journal-title":"J Clin Med"},{"key":"2026092314110371000_vbag214-B8","doi-asserted-by":"crossref","first-page":"141","DOI":"10.1038\/s43856-023-00370-1","article-title":"The future landscape of large language models in medicine","volume":"3","author":"Clusmann","year":"2023","journal-title":"Commun Med (Lond)"},{"key":"2026092314110371000_vbag214-B9","doi-asserted-by":"crossref","first-page":"vbad006","DOI":"10.1093\/bioadv\/vbad006","article-title":"Pancancer survival prediction using a deep learning architecture with multimodal representation and integration","volume":"3","author":"Fan","year":"2023","journal-title":"Bioinform Adv"},{"key":"2026092314110371000_vbag214-B10","doi-asserted-by":"crossref","first-page":"802","DOI":"10.1001\/jama.1996.03540100046027","article-title":"The logistic organ dysfunction system. A new way to assess organ dysfunction in the intensive care unit. ICU Scoring Group","volume":"276","author":"Gall","year":"1996","journal-title":"JAMA"},{"key":"2026092314110371000_vbag214-B11","doi-asserted-by":"publisher","first-page":"62","DOI":"10.1007\/978-981-96-7005-5_5","volume-title":"Neural Information Processing","author":"Gana","year":"2025"},{"key":"2026092314110371000_vbag214-B12","first-page":"e215","volume-title":". Circulation","author":"Goldberger","year":"2000"},{"key":"2026092314110371000_vbag214-B13","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1186\/s13613-023-01102-3","article-title":"Critically ill patient mortality by age: long-term follow-up (CIMbA-LT)","volume":"13","author":"Gon\u00e7alves-Pereira","year":"2023","journal-title":"Ann Intensive Care"},{"key":"2026092314110371000_vbag214-B14","author":"Grnarova","year":"2016"},{"key":"2026092314110371000_vbag214-B15","doi-asserted-by":"crossref","first-page":"vbaf121","DOI":"10.1093\/bioadv\/vbaf121","article-title":"Predictive machine learning model for 30-day hospital readmissions in a tertiary healthcare setting","volume":"5","author":"Halac","year":"2025","journal-title":"Bioinform Adv"},{"key":"2026092314110371000_vbag214-B16","doi-asserted-by":"crossref","first-page":"6930","DOI":"10.3390\/app13126930","article-title":"Prediction of intensive care unit length of stay in the mimic-iv dataset","volume":"13","author":"Hempel","year":"2023","journal-title":"Appl Sci"},{"key":"2026092314110371000_vbag214-B17","doi-asserted-by":"crossref","first-page":"462","DOI":"10.1186\/s12967-020-02620-5","article-title":"Predicting 30-days mortality for MIMIC-III patients with sepsis-3: a machine learning approach using XGboost","volume":"18","author":"Hou","year":"2020","journal-title":"J Transl Med"},{"key":"2026092314110371000_vbag214-B18","doi-asserted-by":"publisher","first-page":"136","DOI":"10.1038\/s41746-020-00341-z","article-title":"Fusion of medical imaging and electronic health records using deep learning: a systematic review and implementation guidelines","volume":"3","author":"Huang","year":"2020","journal-title":"NPJ Digit Med"},{"key":"2026092314110371000_vbag214-B19","author":"Johnson","year":"2024"},{"key":"2026092314110371000_vbag214-B20","doi-asserted-by":"crossref","first-page":"219","DOI":"10.1038\/s41597-023-02136-9","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci Data"},{"key":"2026092314110371000_vbag214-B21","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1093\/jamia\/ocx084","article-title":"The MIMIC code repository: enabling reproducibility in critical care research","volume":"25","author":"Johnson","year":"2018","journal-title":"J Am Med Inform Assoc"},{"key":"2026092314110371000_vbag214-B22","first-page":"1","author":"Kollapally","year":"2023"},{"key":"2026092314110371000_vbag214-B23","author":"Kossen","year":"2021"},{"key":"2026092314110371000_vbag214-B24","doi-asserted-by":"publisher","author":"Kweon","year":"2024","DOI":"10.13026\/ckbf-pt70"},{"key":"2026092314110371000_vbag214-B25","doi-asserted-by":"publisher","author":"Lam","year":"2025","DOI":"10.1016\/j.rpth.2025.102896"},{"key":"2026092314110371000_vbag214-B26","doi-asserted-by":"crossref","first-page":"2957","DOI":"10.1001\/jama.1993.03510240069035","article-title":"A new simplified acute physiology score (SAPS II) based on a European\/North American multicenter study","volume":"270","author":"Le Gall","year":"1993","journal-title":"JAMA"},{"key":"2026092314110371000_vbag214-B27","doi-asserted-by":"crossref","first-page":"R645","DOI":"10.1186\/cc3821","article-title":"Mortality prediction using SAPS II: an update for French intensive care units","volume":"9","author":"Le Gall","year":"2005","journal-title":"Crit Care"},{"key":"2026092314110371000_vbag214-B28","doi-asserted-by":"crossref","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":"2026092314110371000_vbag214-B29","doi-asserted-by":"crossref","first-page":"100511","DOI":"10.1016\/j.cosrev.2022.100511","article-title":"Neural natural language processing for unstructured data in electronic health records: a review","volume":"46","author":"Li","year":"2022","journal-title":"Comput Sci Rev"},{"key":"2026092314110371000_vbag214-B30","first-page":"10632","author":"Li","year":"2024"},{"key":"2026092314110371000_vbag214-B31","doi-asserted-by":"crossref","first-page":"vbac065","DOI":"10.1093\/bioadv\/vbac065","article-title":"Integrating multimodal data through interpretable heterogeneous ensembles","volume":"2","author":"Li","year":"2022","journal-title":"Bioinform Adv"},{"key":"2026092314110371000_vbag214-B32","doi-asserted-by":"crossref","first-page":"e31251","DOI":"10.1097\/MD.0000000000031251","article-title":"Predicting in-hospital mortality for MIMIC-III patients: a nomogram combined with sofa score","volume":"101","author":"Liu","year":"2022","journal-title":"Medicine (Baltimore)"},{"key":"2026092314110371000_vbag214-B33","first-page":"189","author":"Lovon-Melgarejo","year":"2024"},{"key":"2026092314110371000_vbag214-B34","article-title":"A unified approach to interpreting model predictions","author":"Lundberg","journal-title":"Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS 2017)"},{"key":"2026092314110371000_vbag214-B35","first-page":"719","article-title":"A multimodal transformer: fusing clinical notes with structured ehr data for interpretable in-hospital mortality prediction","volume":"2022","author":"Lyu","year":"2023","journal-title":"AMIA Annu Symp Proc"},{"key":"2026092314110371000_vbag214-B37","author":"MIT Laboratory for Computational Physiology","year":"2019"},{"key":"2026092314110371000_vbag214-B38","doi-asserted-by":"crossref","first-page":"210","DOI":"10.7326\/M23-2772","article-title":"Large language models in medicine: the potentials and pitfalls: a narrative review","volume":"177","author":"Omiye","year":"2024","journal-title":"Ann Intern Med"},{"key":"2026092314110371000_vbag214-B39","doi-asserted-by":"crossref","first-page":"e118","DOI":"10.14440\/jbm.2019.299","article-title":"FairSubset: a tool to choose representative subsets of data for use with replicates or groups of different sample sizes","volume":"6","author":"Ortell","year":"2019","journal-title":"J Biol Methods"},{"key":"2026092314110371000_vbag214-B40","doi-asserted-by":"crossref","first-page":"1130","DOI":"10.1097\/01.mlr.0000182534.19832.83","article-title":"Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data","volume":"43","author":"Quan","year":"2005","journal-title":"Med Care"},{"key":"2026092314110371000_vbag214-B41","first-page":"63","volume-title":"International Symposium on Distributed Computing and Artificial Intelligence","author":"Rubi\u00f1os","year":"2024"},{"key":"2026092314110371000_vbag214-B36","first-page":"1","article-title":"Machine learning for predicting post-partum anemia: a comparative performance analysis","volume":"2025","journal-title":"IEEE J Biomed Health Inform"},{"key":"2026092314110371000_vbag214-B42","doi-asserted-by":"crossref","first-page":"1589","DOI":"10.1109\/JBHI.2017.2767063","article-title":"Deep EHR: a survey of recent advances in deep learning techniques for electronic health record (EHR) analysis","volume":"22","author":"Shickel","year":"2018","journal-title":"IEEE J Biomed Health Inform"},{"key":"2026092314110371000_vbag214-B43","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1038\/s41746-022-00689-4","article-title":"Integrated multimodal artificial intelligence framework for healthcare applications","volume":"5","author":"Soenksen","year":"2022","journal-title":"NPJ Digit Med"},{"key":"2026092314110371000_vbag214-B44","doi-asserted-by":"crossref","first-page":"100173","DOI":"10.1016\/j.teler.2024.100173","article-title":"Advancements in natural language processing: implications, challenges, and future directions","volume":"16","author":"Wibawa","year":"2024","journal-title":"Telematics Inform Rep"},{"key":"2026092314110371000_vbag214-B45","doi-asserted-by":"crossref","first-page":"bbad493","DOI":"10.1093\/bib\/bbad493","article-title":"Opportunities and challenges for ChatGPT and large language models in biomedicine and health","volume":"25","author":"Tian","year":"2023","journal-title":"Brief Bioinform"},{"key":"2026092314110371000_vbag214-B46","doi-asserted-by":"crossref","first-page":"91","DOI":"10.1186\/1471-2105-7-91","article-title":"Bias in error estimation when using cross-validation for model selection","volume":"7","author":"Varma","year":"2006","journal-title":"BMC Bioinformatics"},{"key":"2026092314110371000_vbag214-B47","doi-asserted-by":"crossref","first-page":"707","DOI":"10.1007\/BF01709751","article-title":"The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction\/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine","volume":"22","author":"Vincent","year":"1996","journal-title":"Intensive Care Med"},{"key":"2026092314110371000_vbag214-B48","doi-asserted-by":"crossref","first-page":"353","DOI":"10.1016\/S0749-0704(05)70114-7","article-title":"Scoring systems for assessing organ dysfunction and survival","volume":"16","author":"Vincent","year":"2000","journal-title":"Crit Care Clin"},{"key":"2026092314110371000_vbag214-B49","doi-asserted-by":"crossref","first-page":"232","DOI":"10.1016\/j.neucom.2015.08.104","article-title":"Auto-encoder based dimensionality reduction","volume":"184","author":"Wang","year":"2016","journal-title":"Neurocomputing (Amst)"},{"key":"2026092314110371000_vbag214-B50","doi-asserted-by":"crossref","first-page":"7546","DOI":"10.3390\/app14177546","article-title":"Integrating structured and unstructured data with BERTopic and machine learning: a comprehensive predictive model for mortality in ICU heart failure patients","volume":"14","author":"Wu","year":"2024","journal-title":"Appl Sci"},{"key":"2026092314110371000_vbag214-B51","doi-asserted-by":"crossref","first-page":"2625","DOI":"10.2147\/JMDH.S416943","article-title":"A retrospective cohort study: predicting 90-day mortality for ICU trauma patients with a machine learning algorithm using XGboost using MIMIC-III database","volume":"16","author":"Yang","year":"2023","journal-title":"J Multidiscip Healthc"},{"key":"2026092314110371000_vbag214-B52","doi-asserted-by":"publisher","author":"Zafarmomen","year":"2025","DOI":"10.1016\/j.envsoft.2025.106421"},{"key":"2026092314110371000_vbag214-B53","doi-asserted-by":"crossref","first-page":"e59069","DOI":"10.2196\/59069","article-title":"Revolutionizing health care: the transformative impact of large language models in medicine","volume":"27","author":"Zhang","year":"2025","journal-title":"J Med Internet Res"},{"key":"2026092314110371000_vbag214-B54","doi-asserted-by":"crossref","first-page":"372","DOI":"10.1186\/s12911-024-02793-9","article-title":"Integrating structured and unstructured data for predicting emergency severity: an association and predictive study using transformer-based natural language processing models","volume":"24","author":"Zhang","year":"2024","journal-title":"BMC Med Inform Decis Mak"}],"container-title":["Bioinformatics Advances"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/advance-article-pdf\/doi\/10.1093\/bioadv\/vbag214\/70496294\/vbag214.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/article-pdf\/6\/1\/vbag214\/70496294\/vbag214.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/article-pdf\/6\/1\/vbag214\/70496294\/vbag214.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T18:11:12Z","timestamp":1790187072000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformaticsadvances\/article\/doi\/10.1093\/bioadv\/vbag214\/8752612"}},"subtitle":[],"editor":[{"given":"Mira","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":54,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1,7]]}},"URL":"https:\/\/doi.org\/10.1093\/bioadv\/vbag214","relation":{},"ISSN":["2635-0041"],"issn-type":[{"value":"2635-0041","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026]]},"published":{"date-parts":[[2026]]},"article-number":"vbag214"}}