Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366824871> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W4366824871 endingPage "104370" @default.
- W4366824871 startingPage "104370" @default.
- W4366824871 abstract "To develop a natural language processing (NLP) system to extract medications and contextual information that help understand drug changes. This project is part of the 2022 n2c2 challenge. We developed NLP systems for medication mention extraction, event classification (indicating medication changes discussed or not), and context classification to classify medication changes context into 5 orthogonal dimensions related to drug changes. We explored 6 state-of-the-art pretrained transformer models for the three subtasks, including GatorTron, a large language model pretrained using > 90 billion words of text (including > 80 billion words from > 290 million clinical notes identified at the University of Florida Health). We evaluated our NLP systems using annotated data and evaluation scripts provided by the 2022 n2c2 organizers. Our GatorTron models achieved the best F1-scores of 0.9828 for medication extraction (ranked 3rd), 0.9379 for event classification (ranked 2nd), and the best micro-average accuracy of 0.9126 for context classification. GatorTron outperformed existing transformer models pretrained using smaller general English text and clinical text corpora, indicating the advantage of large language models. This study demonstrated the advantage of using large transformer models for contextual medication information extraction from clinical narratives." @default.
- W4366824871 created "2023-04-25" @default.
- W4366824871 creator A5004483523 @default.
- W4366824871 creator A5010253402 @default.
- W4366824871 creator A5030951014 @default.
- W4366824871 creator A5052505769 @default.
- W4366824871 creator A5067895217 @default.
- W4366824871 creator A5086486425 @default.
- W4366824871 date "2023-06-01" @default.
- W4366824871 modified "2023-10-18" @default.
- W4366824871 title "Contextualized medication information extraction using Transformer-based deep learning architectures" @default.
- W4366824871 cites W2104381725 @default.
- W4366824871 cites W2114039834 @default.
- W4366824871 cites W2137407193 @default.
- W4366824871 cites W2159636537 @default.
- W4366824871 cites W2162364899 @default.
- W4366824871 cites W2168041406 @default.
- W4366824871 cites W2908038241 @default.
- W4366824871 cites W2911489562 @default.
- W4366824871 cites W2947903144 @default.
- W4366824871 cites W2963716420 @default.
- W4366824871 cites W2970198438 @default.
- W4366824871 cites W2979250794 @default.
- W4366824871 cites W3080429061 @default.
- W4366824871 cites W3095092693 @default.
- W4366824871 cites W3111654544 @default.
- W4366824871 cites W3148040514 @default.
- W4366824871 cites W4312220150 @default.
- W4366824871 doi "https://doi.org/10.1016/j.jbi.2023.104370" @default.
- W4366824871 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37100106" @default.
- W4366824871 hasPublicationYear "2023" @default.
- W4366824871 type Work @default.
- W4366824871 citedByCount "1" @default.
- W4366824871 countsByYear W43668248712023 @default.
- W4366824871 crossrefType "journal-article" @default.
- W4366824871 hasAuthorship W4366824871A5004483523 @default.
- W4366824871 hasAuthorship W4366824871A5010253402 @default.
- W4366824871 hasAuthorship W4366824871A5030951014 @default.
- W4366824871 hasAuthorship W4366824871A5052505769 @default.
- W4366824871 hasAuthorship W4366824871A5067895217 @default.
- W4366824871 hasAuthorship W4366824871A5086486425 @default.
- W4366824871 hasBestOaLocation W43668248712 @default.
- W4366824871 hasConcept C111919701 @default.
- W4366824871 hasConcept C119599485 @default.
- W4366824871 hasConcept C119857082 @default.
- W4366824871 hasConcept C127413603 @default.
- W4366824871 hasConcept C137293760 @default.
- W4366824871 hasConcept C148524875 @default.
- W4366824871 hasConcept C154945302 @default.
- W4366824871 hasConcept C165801399 @default.
- W4366824871 hasConcept C195807954 @default.
- W4366824871 hasConcept C204321447 @default.
- W4366824871 hasConcept C41008148 @default.
- W4366824871 hasConcept C61423126 @default.
- W4366824871 hasConcept C66322947 @default.
- W4366824871 hasConceptScore W4366824871C111919701 @default.
- W4366824871 hasConceptScore W4366824871C119599485 @default.
- W4366824871 hasConceptScore W4366824871C119857082 @default.
- W4366824871 hasConceptScore W4366824871C127413603 @default.
- W4366824871 hasConceptScore W4366824871C137293760 @default.
- W4366824871 hasConceptScore W4366824871C148524875 @default.
- W4366824871 hasConceptScore W4366824871C154945302 @default.
- W4366824871 hasConceptScore W4366824871C165801399 @default.
- W4366824871 hasConceptScore W4366824871C195807954 @default.
- W4366824871 hasConceptScore W4366824871C204321447 @default.
- W4366824871 hasConceptScore W4366824871C41008148 @default.
- W4366824871 hasConceptScore W4366824871C61423126 @default.
- W4366824871 hasConceptScore W4366824871C66322947 @default.
- W4366824871 hasFunder F4320308927 @default.
- W4366824871 hasFunder F4320309480 @default.
- W4366824871 hasFunder F4320332162 @default.
- W4366824871 hasFunder F4320337337 @default.
- W4366824871 hasFunder F4320337346 @default.
- W4366824871 hasFunder F4320337347 @default.
- W4366824871 hasFunder F4320337351 @default.
- W4366824871 hasLocation W43668248711 @default.
- W4366824871 hasLocation W43668248712 @default.
- W4366824871 hasLocation W43668248713 @default.
- W4366824871 hasLocation W43668248714 @default.
- W4366824871 hasOpenAccess W4366824871 @default.
- W4366824871 hasPrimaryLocation W43668248711 @default.
- W4366824871 hasRelatedWork W2936497627 @default.
- W4366824871 hasRelatedWork W3013624417 @default.
- W4366824871 hasRelatedWork W3049463507 @default.
- W4366824871 hasRelatedWork W3094871513 @default.
- W4366824871 hasRelatedWork W3098382480 @default.
- W4366824871 hasRelatedWork W3126642501 @default.
- W4366824871 hasRelatedWork W4287598411 @default.
- W4366824871 hasRelatedWork W4287826556 @default.
- W4366824871 hasRelatedWork W4288267738 @default.
- W4366824871 hasRelatedWork W4288365749 @default.
- W4366824871 hasVolume "142" @default.
- W4366824871 isParatext "false" @default.
- W4366824871 isRetracted "false" @default.
- W4366824871 workType "article" @default.