{"id":"W3197008538","doi":"10.1038/s41467-021-25578-4","title":"Automatically disambiguating medical acronyms with ontology-aware deep learning","year":2021,"lang":"en","type":"article","venue":"Nature Communications","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; Hospital for Sick Children; Toronto Rehabilitation Institute; University of Toronto; University Health Network","funders":"Hospital for Sick Children","keywords":"Computer science; Workflow; Task (project management); Context (archaeology); Artificial intelligence; Ontology; Training set; Set (abstract data type); Test set; Labeled data; Natural language processing; Machine learning; Information retrieval; Database","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001128035,0.001400484,0.0006682602,0.003923655,0.0006805343,0.001327762,0.001534178,0.001265017,0.003036016],"category_scores_gemma":[0.003974604,0.0003969222,0.001216599,0.00252773,0.0006976699,0.002411602,0.002113113,0.001945784,0.002817655],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001376927,"about_ca_system_score_gemma":0.002414478,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008074093,"about_ca_topic_score_gemma":0.01764008,"domain_scores_codex":[0.9990516,0.0001810624,0.0001105752,0.0004192299,0.0001635215,0.0000740813],"domain_scores_gemma":[0.9983838,0.0007172533,0.0002243053,0.0002903857,0.0002963209,0.0000878885],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004620689,0.0003431804,0.01156058,0.0007580108,0.0001913523,0.0006968727,0.0004427221,0.03971218,0.01989491,0.009783482,0.0954554,0.8206992],"study_design_scores_gemma":[0.0001543671,0.0001483556,0.00624016,0.0003020645,0.0002141179,0.0008766626,0.0005810678,0.8180265,0.02769422,0.04653297,0.09912572,0.0001038056],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1654275,0.008323068,0.7392828,0.004650509,0.00196154,0.0004966545,0.02380896,0.03909499,0.01695397],"genre_scores_gemma":[0.4218783,0.001929073,0.5191171,0.001715486,0.0004478048,0.0002672779,0.04521511,0.0006698795,0.00875987],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008074093,"threshold_uncertainty_score":0.01605421,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01294677349322768,"score_gpt":0.3175330017944535,"score_spread":0.3045862283012258,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}