{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003029172,0.001324374,0.001040795,0.004529414,0.0005494585,0.001878174,0.002193549,0.001943461,0.001270576],"category_scores_gemma":[0.01160949,0.000726119,0.002526303,0.004374539,0.0005925077,0.002668044,0.001776137,0.002206989,0.001451302],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009810579,"about_ca_system_score_gemma":0.001349405,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006329267,"about_ca_topic_score_gemma":0.01186064,"domain_scores_codex":[0.9973432,0.0007951038,0.0003324191,0.0009857325,0.0003398456,0.0002037189],"domain_scores_gemma":[0.9898726,0.007457007,0.001056668,0.0007633379,0.0006448435,0.0002054479],"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":[0.001681256,0.00109757,0.1159285,0.002277503,0.001201095,0.004087056,0.003529824,0.1696705,0.01875988,0.02106294,0.05728543,0.6034184],"study_design_scores_gemma":[0.0001377382,0.0001088081,0.01177,0.0001931284,0.0002539189,0.0007969422,0.0006114932,0.9277411,0.003267379,0.04329072,0.01173305,0.0000957406],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09125872,0.005737207,0.8750171,0.002835816,0.000327445,0.0004402886,0.01972364,0.003021192,0.001638686],"genre_scores_gemma":[0.6526348,0.003387906,0.2875519,0.0009012012,0.001070557,0.0008860055,0.04936134,0.0002971657,0.003909176],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006329267,"threshold_uncertainty_score":0.01601994,"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."}}