{"id":"W3210129127","doi":"10.1186/s12874-021-01434-3","title":"Standardizing registry data to the OMOP Common Data Model: experience from three pulmonary hypertension databases","year":2021,"lang":"en","type":"article","venue":"BMC Medical Research Methodology","topic":"Pulmonary Hypertension Research and Treatments","field":"Medicine","cited_by":86,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Actelion Pharmaceuticals","keywords":"Observational study; Computer science; Missing data; Data mining; Medicine; Information retrieval; Database; Data science; Machine learning; Internal medicine","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","open_science","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.01757702,0.00030559,0.00100704,0.0002089399,0.0006426068,0.00006861159,0.003864666,0.0002852183,0.001237527],"category_scores_gemma":[0.1611974,0.0001910108,0.0001077171,0.0009746409,0.001063072,0.0003547292,0.0144013,0.001907735,0.0001886761],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001202853,"about_ca_system_score_gemma":0.005185454,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003372167,"about_ca_topic_score_gemma":0.007080073,"domain_scores_codex":[0.9828846,0.007741921,0.0006805545,0.002189509,0.005112287,0.001391106],"domain_scores_gemma":[0.9408724,0.04075246,0.0001071876,0.01444341,0.001123367,0.002701229],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.01084291,0.003654865,0.07804661,0.0004915498,0.001055813,0.1584607,0.0003736687,0.00007163893,0.05702934,0.000883274,0.3270312,0.3620584],"study_design_scores_gemma":[0.00428958,0.000658266,0.03890277,0.001818648,0.0003652186,0.008926157,0.005200589,0.655687,0.004093091,0.00296359,0.2763832,0.0007119108],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5611498,0.02818884,0.3276821,0.0716624,0.000843057,0.002136259,0.006195426,0.0002213842,0.001920761],"genre_scores_gemma":[0.4451143,0.008460587,0.514311,0.008299652,0.001629075,0.0002504164,0.02015682,0.0001461408,0.001632085],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6556154,"threshold_uncertainty_score":0.9996755,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.8345554526349793,"score_gpt":0.5887189182897148,"score_spread":0.2458365343452645,"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."}}