{"id":"W2940730617","doi":"10.2196/11499","title":"Adapting State-of-the-Art Deep Language Models to Clinical Information Extraction Systems: Potentials, Challenges, and Solutions","year":2019,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Department of Education and Training; Australian National University","keywords":"Computer science; Artificial intelligence; Natural language processing; Information extraction; Vocabulary; Context (archaeology); Health informatics; Informatics; Natural language; Machine learning; Data science; Health care; Human–computer interaction","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006103388,0.001138089,0.0006872395,0.00103642,0.0003192298,0.00209987,0.002518875,0.001909693,0.00193195],"category_scores_gemma":[0.01355305,0.0006036993,0.0011002,0.001118256,0.00104236,0.004040662,0.002186297,0.002677348,0.00122064],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001464976,"about_ca_system_score_gemma":0.001874628,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006263972,"about_ca_topic_score_gemma":0.005225802,"domain_scores_codex":[0.9978497,0.001030743,0.0001916067,0.00041871,0.0003422571,0.0001669847],"domain_scores_gemma":[0.9938122,0.003825539,0.0002616998,0.0009058928,0.001031076,0.0001636322],"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":[0.0002913504,0.0004938957,0.006040737,0.0006973973,0.0002474424,0.0002594715,0.0005387882,0.1964536,0.01411742,0.008515953,0.007736362,0.7646075],"study_design_scores_gemma":[0.00002807491,0.0001877101,0.0008725697,0.0001149694,0.00005721074,0.0001024635,0.000197322,0.9629373,0.01122831,0.01689247,0.007340917,0.00004068794],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09093706,0.01229078,0.8723794,0.01272335,0.0004636964,0.000290881,0.0005060402,0.005315078,0.005093704],"genre_scores_gemma":[0.5877813,0.007847837,0.3953209,0.002459496,0.0003598879,0.0003580906,0.001512484,0.0003125866,0.00404745],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006263972,"threshold_uncertainty_score":0.03227818,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05289056690827042,"score_gpt":0.3127531179032773,"score_spread":0.2598625509950069,"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."}}