{"id":"W3106195149","doi":"10.1093/jamia/ocaa225","title":"Assessing the quality of clinical and administrative data extracted from hospitals: the General Medicine Inpatient Initiative (GEMINI) experience","year":2020,"lang":"en","type":"article","venue":"Journal of the American Medical Informatics Association","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":96,"is_retracted":false,"has_abstract":true,"ca_institutions":"Sunnybrook Health Science Centre; Trillium Health Centre; Institute for Clinical Evaluative Sciences; Health Sciences Centre; Mount Sinai Hospital; University of Toronto; University Health Network; Queen's University; St. Michael's Hospital","funders":"University of Toronto","keywords":"Data quality; Data extraction; Computer science; Gold standard (test); Quality (philosophy); Quality management; Data collection; Data mining; Database; Medicine; MEDLINE; Statistics; Operations management; Mathematics","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.08441567,0.0006869121,0.0007062837,0.003570054,0.00129702,0.004396054,0.002743778,0.0007317749,0.0007157022],"category_scores_gemma":[0.229331,0.0005387535,0.001086156,0.009297737,0.002052467,0.00217478,0.003791917,0.001181521,0.0001831384],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.01594896,"about_ca_system_score_gemma":0.02054217,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.2271769,"about_ca_topic_score_gemma":0.2214451,"domain_scores_codex":[0.9433017,0.02632327,0.007050904,0.004033037,0.01799023,0.001300948],"domain_scores_gemma":[0.7605157,0.1266459,0.02672993,0.02601043,0.05619465,0.003903469],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0007047969,0.0001607664,0.8687367,0.0009049939,0.0006692078,0.0001614524,0.004858674,0.01252732,0.001221964,0.003288073,0.01715983,0.08960614],"study_design_scores_gemma":[0.0003277104,0.0006826301,0.8505705,0.001070365,0.0005249447,0.000631149,0.005413148,0.07881287,0.007263667,0.004621894,0.04983516,0.0002459465],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9123253,0.003021828,0.03961053,0.01317732,0.0002079769,0.001438264,0.02064169,0.001386679,0.008190529],"genre_scores_gemma":[0.8826638,0.0008779062,0.08732111,0.001633627,0.0001142742,0.0003089151,0.02614539,0.0002270659,0.0007078869],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9155843,"threshold_uncertainty_score":0.4517092,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.540661436719735,"score_gpt":0.5948575964308802,"score_spread":0.05419615971114522,"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."}}