{"id":"W3011721738","doi":"10.1101/2020.03.16.20036962","title":"Assessing the quality of clinical and administrative data extracted from hospitals: The General Medicine Inpatient Initiative (GEMINI) experience","year":2020,"lang":"en","type":"preprint","venue":"medRxiv","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"Sunnybrook Health Science Centre; Trillium Health Centre; Institute for Clinical Evaluative Sciences; University Health Network; University of Toronto; Health Sciences Centre; Mount Sinai Hospital; Queen's University; St. Michael's Hospital","funders":"","keywords":"Data quality; Gold standard (test); Computer science; Medicine; Quality (philosophy); Quality management; Medical record; Data mining; Database; Missing data; Data collection; Medical physics; Statistics; Machine learning; Operations management; Surgery; Internal medicine; Mathematics","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.1259025,0.0007010163,0.0008961359,0.003616771,0.001157924,0.004226411,0.003145333,0.0009982294,0.0007467061],"category_scores_gemma":[0.252673,0.0006401864,0.001022592,0.008205847,0.002187168,0.002372542,0.004786927,0.001393171,0.0002915593],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.009395327,"about_ca_system_score_gemma":0.01041858,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.08794669,"about_ca_topic_score_gemma":0.07895877,"domain_scores_codex":[0.9270381,0.04101029,0.008049775,0.005554864,0.01656198,0.001784973],"domain_scores_gemma":[0.6959934,0.1648632,0.0257646,0.04363737,0.06397807,0.005763409],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001256083,0.0003539211,0.8528337,0.0008685919,0.0006341605,0.0002050691,0.006788601,0.008416928,0.001558068,0.002604315,0.01926638,0.1052142],"study_design_scores_gemma":[0.0004814394,0.001108115,0.8705078,0.0009860486,0.0006087123,0.0007337841,0.005671756,0.05673925,0.009754553,0.003419993,0.04968333,0.0003052106],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9397043,0.0018556,0.0246559,0.006706051,0.0001871934,0.001577988,0.01754601,0.00146326,0.00630373],"genre_scores_gemma":[0.8919343,0.0005625854,0.07344418,0.001687967,0.0001303642,0.0004546354,0.03076459,0.0003165787,0.0007048573],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8740975,"threshold_uncertainty_score":0.665844,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8098649904626021,"score_gpt":0.6374224943190503,"score_spread":0.1724424961435518,"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."}}