{"id":"W2144316471","doi":"10.1136/amiajnl-2014-002768","title":"A novel method of adverse event detection can accurately identify venous thromboembolisms (VTEs) from narrative electronic health record data","year":2014,"lang":"en","type":"article","venue":"Journal of the American Medical Informatics Association","topic":"Venous Thromboembolism Diagnosis and Management","field":"Medicine","cited_by":82,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University Health Centre; McGill University","funders":"Canadian Institutes of Health Research","keywords":"Medicine; Pulmonary embolism; Deep vein; Internal medicine; Electronic health record; Venous thrombosis; Receiver operating characteristic; Venous thromboembolism; Thrombosis; Predictive value; Surgery; Health care","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004728594,0.0001699454,0.001144981,0.0001538801,0.0001436465,0.00001935525,0.0006158563,0.00008004696,0.00004234455],"category_scores_gemma":[0.002769604,0.0001179813,0.0002324438,0.000471297,0.00008021492,0.0002811915,0.0002232783,0.0007102473,0.000004130732],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00111101,"about_ca_system_score_gemma":0.0009185312,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002369262,"about_ca_topic_score_gemma":0.0008777842,"domain_scores_codex":[0.9958609,0.0003697433,0.001601766,0.000118866,0.001678289,0.0003703777],"domain_scores_gemma":[0.9919237,0.0004639809,0.006469456,0.00050719,0.0003879879,0.0002477041],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0004120515,0.003307491,0.006659296,0.0003881393,0.005078703,0.000004627832,0.03392285,0.0008025033,0.00357729,0.0004475455,0.01762356,0.9277759],"study_design_scores_gemma":[0.01128766,0.01084348,0.7649856,0.002878113,0.003693543,0.0003145495,0.04042632,0.1261698,0.004724743,0.003288944,0.03049888,0.0008884096],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7463059,0.00005505733,0.2437844,0.008455181,0.0008090656,0.0004221556,0.00004342093,0.00001601494,0.000108768],"genre_scores_gemma":[0.9844065,0.0006175133,0.009618909,0.004915672,0.0003666477,0.000004975563,0.00003524181,0.00001870267,0.00001589627],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9268875,"threshold_uncertainty_score":0.4811139,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02710845640594695,"score_gpt":0.3615146955703141,"score_spread":0.3344062391643672,"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."}}