{"id":"W4377292073","doi":"10.1136/bmjoq-2022-002119","title":"Implementation of the YEARS algorithm to optimise pulmonary embolism diagnostic workup in the emergency department","year":2023,"lang":"en","type":"article","venue":"BMJ Open Quality","topic":"Venous Thromboembolism Diagnosis and Management","field":"Medicine","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; Health Sciences Centre; Sunnybrook Health Science Centre; University of Toronto","funders":"","keywords":"Emergency department; Pulmonary embolism; Medicine; Algorithm; Medical emergency; Emergency medicine; Computer science; Cardiology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"observational","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"},{"model":"gpt","categories":[],"domain":null,"study_design":"design_other","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low","status":"direct model label, unvalidated"}],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002535742,0.0001256739,0.0003633591,0.00007617232,0.00008543565,0.00002979089,0.0005534599,0.00003638825,0.0003960893],"category_scores_gemma":[0.0002654853,0.00008182657,0.0001079609,0.0008157941,0.00002649263,0.0000782365,0.0005850167,0.00009780376,0.0001036068],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006335969,"about_ca_system_score_gemma":0.00008293385,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003492472,"about_ca_topic_score_gemma":0.0008770745,"domain_scores_codex":[0.9979205,0.0003700353,0.0006247908,0.0002736346,0.000523142,0.0002878581],"domain_scores_gemma":[0.9987708,0.0002059217,0.0001926295,0.000713219,0.0000473917,0.0000699819],"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":[0.00009015059,0.001771269,0.1549508,0.0002778328,0.0002134794,0.0001660899,0.01042226,0.0003480498,0.0001689398,0.004801366,0.1322057,0.6945841],"study_design_scores_gemma":[0.0005236763,0.00008086032,0.985355,0.00008801652,0.00007356744,0.000002460812,0.005226153,0.00004071934,0.0001093391,0.000494855,0.007885369,0.0001199954],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9771836,0.00005549143,0.00005476302,0.01228124,0.0004088864,0.008133681,0.00005693413,0.00002222868,0.001803169],"genre_scores_gemma":[0.9948143,0.0002656759,0.0004874372,0.001735956,0.0001057215,0.002174584,0.000066537,0.00001766181,0.0003321015],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8304042,"threshold_uncertainty_score":0.5279596,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08688570778566145,"score_gpt":0.4605500231637367,"score_spread":0.3736643153780753,"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."}}