000277551 001__ 277551
000277551 005__ 20250403095900.0
000277551 0247_ $$2doi$$a10.1371/journal.pdig.0000468
000277551 0247_ $$2pmid$$apmid:40100877
000277551 037__ $$aDZNE-2025-00440
000277551 041__ $$aEnglish
000277551 082__ $$a610
000277551 1001_ $$0P:(DE-2719)9002349$$aWegner, Philipp$$b0$$eFirst author$$udzne
000277551 245__ $$aEvaluating knowledge fusion models on detecting adverse drug events in text.
000277551 260__ $$aSan Francisco, CA$$bPLoS$$c2025
000277551 3367_ $$2DRIVER$$aarticle
000277551 3367_ $$2DataCite$$aOutput Types/Journal article
000277551 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1743582837_9485
000277551 3367_ $$2BibTeX$$aARTICLE
000277551 3367_ $$2ORCID$$aJOURNAL_ARTICLE
000277551 3367_ $$00$$2EndNote$$aJournal Article
000277551 520__ $$aDetecting adverse drug events (ADE) of drugs that are already available on the market is an essential part of the pharmacovigilance work conducted by both medical regulatory bodies and the pharmaceutical industry. Concerns regarding drug safety and economic interests serve as motivating factors for the efforts to identify ADEs. Hereby, social media platforms play an important role as a valuable source of reports on ADEs, particularly through collecting posts discussing adverse events associated with specific drugs. We aim with our study to assess the effectiveness of knowledge fusion approaches in combination with transformer-based NLP models to extract ADE mentions from diverse datasets, for instance, texts from Twitter, websites like askapatient.com, and drug labels. The extraction task is formulated as a named entity recognition (NER) problem. The proposed methodology involves applying fusion learning methods to enhance the performance of transformer-based language models with additional contextual knowledge from ontologies or knowledge graphs. Additionally, the study introduces a multi-modal architecture that combines transformer-based language models with graph attention networks (GAT) to identify ADE spans in textual data. A multi-modality model consisting of the ERNIE model with knowledge on drugs reached an F1-score of 71.84% on CADEC corpus. Additionally, a combination of a graph attention network with BERT resulted in an F1-score of 65.16% on SMM4H corpus. Impressively, the same model achieved an F1-score of 72.50% on the PsyTAR corpus, 79.54% on the ADE corpus, and 94.15% on the TAC corpus. Except for the CADEC corpus, the knowledge fusion models consistently outperformed the baseline model, BERT. Our study demonstrates the significance of context knowledge in improving the performance of knowledge fusion models for detecting ADEs from various types of textual data.
000277551 536__ $$0G:(DE-HGF)POF4-353$$a353 - Clinical and Health Care Research (POF4-353)$$cPOF4-353$$fPOF IV$$x0
000277551 588__ $$aDataset connected to CrossRef, PubMed, , Journals: pub.dzne.de
000277551 7001_ $$aFröhlich, Holger$$b1
000277551 7001_ $$00000-0001-9970-4144$$aMadan, Sumit$$b2
000277551 773__ $$0PERI:(DE-600)3106944-7$$a10.1371/journal.pdig.0000468$$gVol. 4, no. 3, p. e0000468 -$$n3$$pe0000468$$tPLOS digital health$$v4$$x2767-3170$$y2025
000277551 8564_ $$uhttps://pub.dzne.de/record/277551/files/DZNE-2025-00440%20SUP.zip
000277551 8564_ $$uhttps://pub.dzne.de/record/277551/files/DZNE-2025-00440.pdf$$yOpenAccess
000277551 8564_ $$uhttps://pub.dzne.de/record/277551/files/DZNE-2025-00440.pdf?subformat=pdfa$$xpdfa$$yOpenAccess
000277551 909CO $$ooai:pub.dzne.de:277551$$popenaire$$popen_access$$pVDB$$pdriver$$pdnbdelivery
000277551 9101_ $$0I:(DE-588)1065079516$$6P:(DE-2719)9002349$$aDeutsches Zentrum für Neurodegenerative Erkrankungen$$b0$$kDZNE
000277551 9131_ $$0G:(DE-HGF)POF4-353$$1G:(DE-HGF)POF4-350$$2G:(DE-HGF)POF4-300$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$aDE-HGF$$bGesundheit$$lNeurodegenerative Diseases$$vClinical and Health Care Research$$x0
000277551 9141_ $$y2025
000277551 915__ $$0StatID:(DE-HGF)0200$$2StatID$$aDBCoverage$$bSCOPUS$$d2025-01-07
000277551 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
000277551 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2024-02-08T09:39:05Z
000277551 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2024-02-08T09:39:05Z
000277551 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000277551 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Anonymous peer review$$d2024-02-08T09:39:05Z
000277551 915__ $$0StatID:(DE-HGF)0561$$2StatID$$aArticle Processing Charges$$d2025-01-07
000277551 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2025-01-07
000277551 915__ $$0StatID:(DE-HGF)0700$$2StatID$$aFees$$d2025-01-07
000277551 9201_ $$0I:(DE-2719)1011401$$kClinical Research Platform (CRP)$$lClinical Research Platform (CRP)$$x0
000277551 980__ $$ajournal
000277551 980__ $$aVDB
000277551 980__ $$aUNRESTRICTED
000277551 980__ $$aI:(DE-2719)1011401
000277551 9801_ $$aFullTexts