Matches in SemOpenAlex for { <https://semopenalex.org/work/W2939470209> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W2939470209 endingPage "122" @default.
- W2939470209 startingPage "116" @default.
- W2939470209 abstract "Electronic medical records (EMRs) are manually annotated by healthcare professionals and specialized medical coders with a standardized set of alphanumeric diagnosis and procedure codes, specifically from the International Classification of Diseases (ICD). Annotating EMRs with ICD codes is important for medical billing and downstream epidemiological studies. However, manually annotating EMRs is both time-consuming and error prone. In this paper, we explore the use of convolutional neural networks (CNNs) for automatic ICD coding. Because many codes occur infrequently, CNN performance is inhibited. Therefore, we propose supplementing EMR data with PubMed indexed biomedical research abstracts through neural transfer learning. Transfer learning is the process of “transferring” knowledge acquired from one task (the source task) to a different (target) task. For the source task, we train a CNN to predict medical subject headings (MeSH) using 1.6 million PubMed indexed biomedical abstracts. For the target task, we train a CNN on 71,463 real-world EMRs collected from the University of Kentucky (UKY) medical center to predict ICD diagnosis codes. We introduce a simple, yet effective, transfer learning methodology which avoids forgetting knowledge gained from the source task. Compared to our prior work using EMRs from the UKY medical center, we improve both the micro and macro F-scores by more than 8%. Likewise, compared to other transfer learning methods, our approach results in nearly 2% improvement in macro F-score. We show that transfer learning can improve CNN performance for EMR coding in the presence of data sparsity issues. Furthermore, we find that our proposed transfer learning approach outperforms other methods with respect to macro F-score. Finally, we analyze how transfer learning impacts codes with respect to code frequency. We find that we achieve greater improvement on infrequent codes compared to improvements in most frequent codes." @default.
- W2939470209 created "2019-04-25" @default.
- W2939470209 creator A5018773113 @default.
- W2939470209 creator A5020481922 @default.
- W2939470209 date "2019-05-01" @default.
- W2939470209 modified "2023-10-08" @default.
- W2939470209 title "Neural transfer learning for assigning diagnosis codes to EMRs" @default.
- W2939470209 cites W1775813496 @default.
- W2939470209 cites W1832693441 @default.
- W2939470209 cites W2096664202 @default.
- W2939470209 cites W2122402213 @default.
- W2939470209 cites W2132755184 @default.
- W2939470209 cites W2149427297 @default.
- W2939470209 cites W2159583324 @default.
- W2939470209 cites W2187523974 @default.
- W2939470209 cites W2396881363 @default.
- W2939470209 cites W2473930607 @default.
- W2939470209 cites W2560647685 @default.
- W2939470209 cites W2615107903 @default.
- W2939470209 cites W2789326915 @default.
- W2939470209 cites W2803920401 @default.
- W2939470209 cites W2951829787 @default.
- W2939470209 cites W2963026768 @default.
- W2939470209 cites W2963042536 @default.
- W2939470209 cites W2963154225 @default.
- W2939470209 doi "https://doi.org/10.1016/j.artmed.2019.04.002" @default.
- W2939470209 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31164204" @default.
- W2939470209 hasPublicationYear "2019" @default.
- W2939470209 type Work @default.
- W2939470209 sameAs 2939470209 @default.
- W2939470209 citedByCount "32" @default.
- W2939470209 countsByYear W29394702092019 @default.
- W2939470209 countsByYear W29394702092020 @default.
- W2939470209 countsByYear W29394702092021 @default.
- W2939470209 countsByYear W29394702092022 @default.
- W2939470209 countsByYear W29394702092023 @default.
- W2939470209 crossrefType "journal-article" @default.
- W2939470209 hasAuthorship W2939470209A5018773113 @default.
- W2939470209 hasAuthorship W2939470209A5020481922 @default.
- W2939470209 hasConcept C105795698 @default.
- W2939470209 hasConcept C108583219 @default.
- W2939470209 hasConcept C119857082 @default.
- W2939470209 hasConcept C150899416 @default.
- W2939470209 hasConcept C154945302 @default.
- W2939470209 hasConcept C162324750 @default.
- W2939470209 hasConcept C179518139 @default.
- W2939470209 hasConcept C187736073 @default.
- W2939470209 hasConcept C199360897 @default.
- W2939470209 hasConcept C2780451532 @default.
- W2939470209 hasConcept C2781003394 @default.
- W2939470209 hasConcept C33923547 @default.
- W2939470209 hasConcept C41008148 @default.
- W2939470209 hasConcept C81363708 @default.
- W2939470209 hasConceptScore W2939470209C105795698 @default.
- W2939470209 hasConceptScore W2939470209C108583219 @default.
- W2939470209 hasConceptScore W2939470209C119857082 @default.
- W2939470209 hasConceptScore W2939470209C150899416 @default.
- W2939470209 hasConceptScore W2939470209C154945302 @default.
- W2939470209 hasConceptScore W2939470209C162324750 @default.
- W2939470209 hasConceptScore W2939470209C179518139 @default.
- W2939470209 hasConceptScore W2939470209C187736073 @default.
- W2939470209 hasConceptScore W2939470209C199360897 @default.
- W2939470209 hasConceptScore W2939470209C2780451532 @default.
- W2939470209 hasConceptScore W2939470209C2781003394 @default.
- W2939470209 hasConceptScore W2939470209C33923547 @default.
- W2939470209 hasConceptScore W2939470209C41008148 @default.
- W2939470209 hasConceptScore W2939470209C81363708 @default.
- W2939470209 hasFunder F4320337372 @default.
- W2939470209 hasLocation W29394702091 @default.
- W2939470209 hasLocation W29394702092 @default.
- W2939470209 hasOpenAccess W2939470209 @default.
- W2939470209 hasPrimaryLocation W29394702091 @default.
- W2939470209 hasRelatedWork W2951211570 @default.
- W2939470209 hasRelatedWork W3133861977 @default.
- W2939470209 hasRelatedWork W3167935049 @default.
- W2939470209 hasRelatedWork W3183901164 @default.
- W2939470209 hasRelatedWork W3192840557 @default.
- W2939470209 hasRelatedWork W3193565141 @default.
- W2939470209 hasRelatedWork W4206357785 @default.
- W2939470209 hasRelatedWork W4226493464 @default.
- W2939470209 hasRelatedWork W4281381188 @default.
- W2939470209 hasRelatedWork W4312417841 @default.
- W2939470209 hasVolume "96" @default.
- W2939470209 isParatext "false" @default.
- W2939470209 isRetracted "false" @default.
- W2939470209 magId "2939470209" @default.
- W2939470209 workType "article" @default.