{"id":"W4294052933","doi":"10.1016/j.jbi.2022.104190","title":"MixEHR-Guided: A guided multi-modal topic modeling approach for large-scale automatic phenotyping using the electronic health record","year":2022,"lang":"en","type":"article","venue":"Journal of Biomedical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University Health Centre; McGill University","funders":"Fonds de recherche du Québec – Nature et technologies; Canadian Institutes of Health Research; Canada First Research Excellence Fund; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Computer science; Machine learning; Artificial intelligence; Health informatics; Data mining; Inference; Population; Data science; Medicine; Public health","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":[0.004418714,0.001502022,0.001544397,0.00180749,0.0007845717,0.001306805,0.00280082,0.001870452,0.002335816],"category_scores_gemma":[0.008199318,0.001082953,0.003010615,0.001346547,0.0008385961,0.001476826,0.002645594,0.002870784,0.001400278],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001104892,"about_ca_system_score_gemma":0.001326619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01200572,"about_ca_topic_score_gemma":0.02003515,"domain_scores_codex":[0.9980028,0.000978799,0.00009259365,0.0006254687,0.0001756171,0.0001247868],"domain_scores_gemma":[0.9938948,0.004983961,0.0002824877,0.0004142487,0.0002723036,0.0001522379],"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":[0.0006956644,0.0004265507,0.01801166,0.000405656,0.0006513087,0.0006985331,0.001697746,0.5312167,0.008309036,0.02877009,0.01958899,0.3895282],"study_design_scores_gemma":[0.00003670921,0.00003031286,0.0006871229,0.00001895914,0.0000319392,0.0000721085,0.00004939393,0.979458,0.0006120056,0.01753012,0.001450723,0.00002258649],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01778434,0.0004977523,0.9764903,0.0006133214,0.00003748858,0.0001276786,0.0009314544,0.003069575,0.000448138],"genre_scores_gemma":[0.3879245,0.0006592947,0.5952135,0.001303694,0.0003608705,0.0009171047,0.00810154,0.001090448,0.004428926],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01200572,"threshold_uncertainty_score":0.02387166,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06237262631338839,"score_gpt":0.3547260243753848,"score_spread":0.2923533980619964,"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."}}