Delirium Identification from Nursing Reports Using Large Language Models

Tietueen suppeat tiedot
dc.contributor.authorGraf, Lisa
dc.contributor.authorRitzi, Alexander
dc.contributor.authorSchöler, Lili M.
dc.contributor.organizationfi=hoitotieteen laitos|en=Department of Nursing Science|
dc.contributor.organization-code1.2.246.10.2458963.20.27201741504
dc.converis.publication-id499069125
dc.converis.urlhttps://research.utu.fi/converis/portal/Publication/499069125
dc.date.accessioned2025-08-27T22:19:28Z
dc.date.available2025-08-27T22:19:28Z
dc.description.abstract<p>This study investigates large language models for delirium detection from nursing reports, comparing keyword matching, prompting, and finetuning. Using a manually labelled dataset from the University Hospital Freiburg, Germany, we tested Llama3 and Phi3 models. Both prompting and finetuning were effective, with finetuning Phi3 (3.8B) achieving the highest accuracy (90.24%) and AUROC (96.07%), significantly outperforming other methods.<br></p>
dc.format.pagerange886
dc.format.pagerange887
dc.identifier.eisbn978-1-64368-596-0
dc.identifier.issn0926-9630
dc.identifier.jour-issn0926-9630
dc.identifier.olddbid201982
dc.identifier.oldhandle10024/185009
dc.identifier.urihttps://www.utupub.fi/handle/11111/39882
dc.identifier.urlhttps://doi.org/10.3233/shti250492
dc.identifier.urnURN:NBN:fi-fe2025082789633
dc.language.isoen
dc.okm.affiliatedauthorSchöler, Lili Maria
dc.okm.discipline316 Nursingen_GB
dc.okm.discipline316 Hoitotiedefi_FI
dc.okm.internationalcopublicationinternational co-publication
dc.okm.internationalityInternational publication
dc.okm.typeA4 Conference Article
dc.publisher.countryNetherlandsen_GB
dc.publisher.countryAlankomaatfi_FI
dc.publisher.country-codeNL
dc.relation.conferenceMedical Informatics Europe Conference
dc.relation.doi10.3233/SHTI250492
dc.relation.ispartofjournalStudies in Health Technology and Informatics
dc.relation.volume327
dc.source.identifierhttps://www.utupub.fi/handle/10024/185009
dc.titleDelirium Identification from Nursing Reports Using Large Language Models
dc.title.bookIntelligent Health Systems – From Technology to Data and Knowledge: Proceedings of MIE 2025
dc.year.issued2025

Tiedostot

Näytetään 1 - 1 / 1
Name:
SHTI-327-SHTI250492(1).pdf
Size:
146.73 KB
Format:
Adobe Portable Document Format
Lataa

Kokoelmat

Rinnakkaistallenteet