{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.203931,0.0006928146,0.0009429682,0.00368591,0.001592842,0.008018436,0.005117459,0.001430483,0.001248608],"category_scores_gemma":[0.3446447,0.0007778303,0.001982213,0.01195437,0.002166749,0.007259482,0.009996469,0.00335441,0.0005962256],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006231084,"about_ca_system_score_gemma":0.01401003,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02209294,"about_ca_topic_score_gemma":0.01036001,"domain_scores_codex":[0.8486201,0.1007159,0.02452574,0.007057526,0.01693844,0.0021424],"domain_scores_gemma":[0.6456518,0.1812595,0.02419239,0.09881069,0.04376717,0.006318462],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.003388323,0.001608039,0.4428138,0.00259009,0.001158949,0.001420264,0.02386218,0.02175297,0.003208281,0.04079985,0.04157869,0.4158187],"study_design_scores_gemma":[0.002372416,0.003417476,0.3581527,0.006662957,0.001877484,0.004157484,0.03864113,0.1216593,0.02905905,0.04998414,0.3831479,0.0008679334],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7191638,0.003207119,0.1981547,0.02335752,0.0006747865,0.006489621,0.03264964,0.005785949,0.01051684],"genre_scores_gemma":[0.606131,0.001606904,0.3256204,0.003086212,0.0001665513,0.003062569,0.05825802,0.001054785,0.001013613],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.796069,"threshold_uncertainty_score":0.9816948,"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."}}