{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.00253066,0.000666902,0.001323702,0.0002869695,0.0003088944,0.0003006146,0.002994087,0.0003115173,0.00002555564],"category_scores_gemma":[0.0010509,0.0006696113,0.00009322212,0.0009028124,0.0001568801,0.0007885603,0.005853248,0.003313502,0.00000237789],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006604984,"about_ca_system_score_gemma":0.002583336,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007255863,"about_ca_topic_score_gemma":0.000503395,"domain_scores_codex":[0.9932835,0.0007753972,0.001108705,0.002795234,0.001034378,0.001002752],"domain_scores_gemma":[0.994632,0.0001870278,0.001447638,0.002977235,0.0004108566,0.000345257],"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":[0.00003633685,0.0001263974,0.9777673,0.003098201,0.0002858141,0.00007046384,0.002972336,0.00003363009,0.0007724843,0.0002888253,0.0003059356,0.01424224],"study_design_scores_gemma":[0.0002684776,0.0001992097,0.327558,0.001100817,0.0001124988,0.00005413094,0.0001246741,0.6694453,0.0001124118,0.0002144111,0.0001519469,0.0006580414],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9351583,0.01108441,0.04494009,0.003346693,0.001064705,0.002700919,0.001138465,0.0005116109,0.00005481885],"genre_scores_gemma":[0.9108115,0.0004635927,0.08629595,0.0002636305,0.0002942141,0.00007055786,0.001691165,0.0000872997,0.00002211411],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6694117,"threshold_uncertainty_score":0.9995755,"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."}}