{"id":"W3085978070","doi":"10.1101/2020.09.11.20192419","title":"Cohort profile: St. Michael’s Hospital Tuberculosis Database (SMH-TB), a retrospective cohort of electronic health record data and variables extracted using natural language processing","year":2020,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; Toronto Metropolitan University; University of Toronto; St. Michael's Hospital","funders":"","keywords":"Medicine; Dictation; Electronic health record; Cohort; Retrospective cohort study; Tuberculosis; Generalizability theory; Medical record; Population; Artificial intelligence; Natural language processing; Database; Pediatrics; Health care; Internal medicine; Statistics; Computer science; Pathology; Speech recognition","routes":{"ca_aff":true,"ca_fund":false,"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":[0.002865822,0.0004310312,0.0004681614,0.002330579,0.0009761982,0.001273817,0.000856449,0.0004685098,0.00470981],"category_scores_gemma":[0.01514029,0.0003119965,0.0006105411,0.004346109,0.000261202,0.000471174,0.0008177641,0.0004937746,0.001565986],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002905805,"about_ca_system_score_gemma":0.007295256,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.35828,"about_ca_topic_score_gemma":0.4395678,"domain_scores_codex":[0.9981559,0.0002309455,0.000388171,0.0005197261,0.0005122959,0.0001930756],"domain_scores_gemma":[0.9878337,0.001599988,0.002180972,0.002486609,0.004833154,0.001065472],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0005897738,0.00008101069,0.9342718,0.0002994988,0.0002465084,0.0003199873,0.0003822755,0.0004065213,0.0008892741,0.000419168,0.05337964,0.008714688],"study_design_scores_gemma":[0.0001600153,0.0001019682,0.9668406,0.0002016599,0.0001808493,0.000620688,0.0005983441,0.001537694,0.000927332,0.0001968669,0.02857966,0.0000544252],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.2717144,0.0003261615,0.002190499,0.0006574402,0.0000611261,0.001025417,0.7215286,0.0002359503,0.002260478],"genre_scores_gemma":[0.3617247,0.0003913395,0.007257133,0.0005844242,0.00008147555,0.001730861,0.626026,0.0001079347,0.002096105],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.35828,"threshold_uncertainty_score":0.7123892,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02011390596484796,"score_gpt":0.3149613491867199,"score_spread":0.294847443221872,"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."}}