{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001366024,0.0001021726,0.0002117375,0.00009425035,0.00006857732,0.00008735577,0.0005179734,0.0001300671,0.00000560268],"category_scores_gemma":[0.0001129842,0.00007775954,0.00005587134,0.0001360519,0.00003746424,0.002608321,0.0004628999,0.0002998784,0.00009281974],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003541812,"about_ca_system_score_gemma":0.000113562,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001888571,"about_ca_topic_score_gemma":0.00001318189,"domain_scores_codex":[0.9977935,0.0000786216,0.00103624,0.00009298311,0.0007558392,0.0002428467],"domain_scores_gemma":[0.9987214,0.0001450347,0.0003266422,0.0004877357,0.0001189999,0.0002002026],"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":[0.000009763127,0.00006605178,0.00007188516,0.0006053747,0.0000357691,0.000001282692,0.04606582,0.01913971,0.00001026395,0.08210238,0.0004967375,0.851395],"study_design_scores_gemma":[0.0002704182,0.00004486,0.0003001147,0.0001803428,0.000004126163,0.00002513239,0.0023464,0.9916533,0.000005207904,0.0003873573,0.004686075,0.0000966359],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07616101,0.0002708375,0.9192549,0.0006264979,0.000865464,0.0005239272,0.000002272045,0.00007968625,0.002215386],"genre_scores_gemma":[0.9685311,0.0002308744,0.03046259,0.0005836896,0.00007610565,0.00003115318,0.000004492859,0.000005528788,0.00007449019],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9725136,"threshold_uncertainty_score":0.3170942,"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."}}