{"id":"W3210677193","doi":"10.1186/s12911-021-01669-6","title":"Assessing the suitability of general practice electronic health records for clinical prediction model development: a data quality assessment","year":2021,"lang":"en","type":"article","venue":"BMC Medical Informatics and Decision Making","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"National Health and Medical Research Council; RACGP Foundation; Medical Research Council; Royal Australian College of General Practitioners; Australian Orthopaedic Association","keywords":"Medicine; Health informatics; Health records; Data quality; Gold standard (test); Medical record; Family medicine; Health care; Public health; Surgery; Operations management; Internal medicine","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.4831795,0.0006827472,0.001662407,0.005325121,0.001420565,0.006716252,0.002631904,0.002027356,0.001182024],"category_scores_gemma":[0.7298711,0.001003301,0.003639862,0.01030522,0.003265823,0.004711732,0.004962535,0.002081122,0.0003025205],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004507841,"about_ca_system_score_gemma":0.007662557,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01027755,"about_ca_topic_score_gemma":0.00809695,"domain_scores_codex":[0.461568,0.38673,0.07737501,0.01074393,0.06136139,0.00222185],"domain_scores_gemma":[0.1125811,0.6760209,0.08756757,0.05522567,0.06690624,0.00169858],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0008696544,0.0002236469,0.9179642,0.002425684,0.0024472,0.0001352658,0.00486392,0.005052791,0.0003599432,0.002318245,0.0021657,0.06117375],"study_design_scores_gemma":[0.0006684516,0.003285565,0.8663828,0.00820033,0.003285589,0.001143987,0.008266825,0.07133432,0.003532592,0.01026409,0.02330223,0.0003332075],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7723535,0.006935596,0.1805626,0.0137291,0.0003852824,0.009130377,0.009664444,0.0004206614,0.006818517],"genre_scores_gemma":[0.9168358,0.0009298949,0.0720327,0.001546561,0.0001477947,0.0033499,0.004864963,0.00007581382,0.0002165836],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5168205,"threshold_uncertainty_score":0.6373318,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2872856093899279,"score_gpt":0.5629759558264738,"score_spread":0.2756903464365458,"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."}}