{"id":"W3144386263","doi":"10.1371/journal.pone.0249622","title":"Mining heterogeneous clinical notes by multi-modal latent topic model","year":2021,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Fonds de recherche du Québec – Nature et technologies; Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Microsoft Research","keywords":"Interpretability; Topic model; Computer science; DECIPHER; Latent Dirichlet allocation; Data science; Natural language processing; Artificial intelligence; Bayesian probability; Information retrieval; Machine learning; Medicine; Bioinformatics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002574132,0.0001306262,0.0002780142,0.00002953617,0.0001099327,0.00008912209,0.0005066534,0.0001164252,0.00002501139],"category_scores_gemma":[0.0005147996,0.0001352682,0.0000829022,0.0001361199,0.00002671889,0.0001058652,0.0004026175,0.0003511656,0.00005957345],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003553195,"about_ca_system_score_gemma":0.0001221948,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003684928,"about_ca_topic_score_gemma":0.00002814556,"domain_scores_codex":[0.9981575,0.0002090256,0.0003908964,0.0005464697,0.0003666003,0.0003295209],"domain_scores_gemma":[0.9986333,0.0002928284,0.00009165148,0.0006591349,0.0001453756,0.0001777185],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002835032,0.007621596,0.8564982,0.0006243125,0.0005379,0.0006013387,0.003491474,0.03039058,0.01366039,0.00216509,0.0007727719,0.08360796],"study_design_scores_gemma":[0.0003056486,0.00007966233,0.002792171,0.00006585079,0.00001299228,0.000008205278,0.000002870699,0.9907375,0.005681908,0.0001077068,0.00005081531,0.0001546738],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8687,0.0005822016,0.1253025,0.004843063,0.0001324187,0.0001183042,0.000004927944,0.0002093619,0.0001072192],"genre_scores_gemma":[0.7142694,0.00004522594,0.2836586,0.001183083,0.00008529897,0.00001208426,0.000008076392,0.0000133614,0.0007248884],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9603469,"threshold_uncertainty_score":0.5516078,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.177569710523654,"score_gpt":0.3476559245245285,"score_spread":0.1700862140008745,"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."}}