{"id":"W2322284403","doi":"10.1016/s1878-7649(14)70145-7","title":"O3.17: Natural language processing of electronic health record data can accurately identify hospitalized elderly patients with venous thromboembolism (VTE)","year":2014,"lang":"en","type":"article","venue":"European Geriatric Medicine","topic":"Venous Thromboembolism Diagnosis and Management","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Medicine; Venous thromboembolism; Electronic health record; Intensive care medicine; Emergency medicine; Medical emergency; Internal medicine; Health care; Thrombosis","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001509022,0.000435469,0.001544548,0.0003262185,0.0001886878,0.00003803808,0.000700769,0.00005176253,0.0001776393],"category_scores_gemma":[0.0003533103,0.0003073658,0.00007127885,0.0007242138,0.0001698375,0.0002048821,0.0002830633,0.0004472901,0.00002727775],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001498789,"about_ca_system_score_gemma":0.0002915085,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009250151,"about_ca_topic_score_gemma":0.0001474435,"domain_scores_codex":[0.9961357,0.0003745825,0.0009443757,0.000831886,0.0008743117,0.0008391248],"domain_scores_gemma":[0.99707,0.0000732468,0.0008813775,0.001396708,0.0002603924,0.0003182355],"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":[0.0006844346,0.00304743,0.01430302,0.002786763,0.0009797548,0.0001975492,0.0113826,0.00001046436,0.0004589925,0.0002342864,0.0978977,0.868017],"study_design_scores_gemma":[0.01116422,0.009800261,0.9453555,0.001452205,0.001084911,0.00003413036,0.0008099725,0.0002812475,0.00003031676,0.00003310296,0.02935949,0.0005946461],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9827607,0.005720349,0.0004716339,0.002655721,0.0006322974,0.001487927,0.00002934383,0.0001966611,0.006045392],"genre_scores_gemma":[0.9943526,0.0006129692,0.0005590016,0.001964337,0.0009356672,0.00001077737,0.001017885,0.0001286811,0.0004180511],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9310525,"threshold_uncertainty_score":0.9999378,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01843880909060731,"score_gpt":0.2984495412215449,"score_spread":0.2800107321309376,"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."}}