{"id":"W4389577465","doi":"10.1109/ichi57859.2023.00103","title":"Leveraging Foundation Models for Clinical Text Analysis","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; Public Health Ontario; University of Toronto","funders":"","keywords":"Computer science; Transformer; Information extraction; Data extraction; Data science; Artificial intelligence; Machine learning; Data mining; Natural language processing; Information retrieval; MEDLINE; Engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007896786,0.00005162847,0.0001251479,0.0001817074,0.00007938365,0.0001163424,0.0003680055,0.00003285719,0.00001454563],"category_scores_gemma":[0.00004669744,0.00004682822,0.0001484987,0.0008199768,0.000007429777,0.0003775133,0.0001351062,0.00004226009,0.00008272553],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001520797,"about_ca_system_score_gemma":0.00002984881,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002492887,"about_ca_topic_score_gemma":0.000007825179,"domain_scores_codex":[0.9990913,0.00002744228,0.0002545849,0.0003284612,0.0001354669,0.0001627077],"domain_scores_gemma":[0.9992353,0.0002202773,0.00004031511,0.0004018339,0.00005932336,0.00004289176],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000002308771,0.00002071967,0.002656647,0.000007386369,0.0001841589,0.00000151072,0.0005076768,0.4043723,0.00001558361,0.2465462,0.001919078,0.3437664],"study_design_scores_gemma":[0.0001112518,0.000007488537,0.0009751967,0.0000011303,0.00002343826,1.594619e-7,0.00001765008,0.9604174,0.00001496703,0.03741964,0.0009505803,0.00006102206],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01322267,0.000004684683,0.9827467,0.001427999,0.0002327362,0.0000902242,2.102354e-7,0.000334874,0.001939951],"genre_scores_gemma":[0.7594512,0.00001093592,0.2381986,0.0003478812,0.00007953406,0.00001493894,0.000006489051,0.000003672784,0.001886781],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7462285,"threshold_uncertainty_score":0.1909599,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2348475168951484,"score_gpt":0.4032111303280295,"score_spread":0.1683636134328811,"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."}}