{"id":"W3135855542","doi":"10.2196/27767","title":"Accuracy of an Artificial Intelligence System for Cancer Clinical Trial Eligibility Screening: Retrospective Pilot Study","year":2021,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":74,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Novartis Pharmaceuticals Corporation","keywords":"Medicine; Breast cancer; Clinical trial; Retrospective cohort study; Inter-rater reliability; Clinical decision support system; Wilcoxon signed-rank test; Medical physics; Cancer; Internal medicine; Artificial intelligence; Statistics; Decision support system; Computer science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003278681,0.0001881644,0.0007887291,0.00009408747,0.0001472077,0.0000392623,0.000241061,0.0002383544,0.0002219579],"category_scores_gemma":[0.007431807,0.0001578041,0.0001716429,0.0004612966,0.0002406601,0.0002776602,0.00007969517,0.0006090493,0.00001765501],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002004491,"about_ca_system_score_gemma":0.00231426,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003986058,"about_ca_topic_score_gemma":0.0004998989,"domain_scores_codex":[0.9948863,0.0002212099,0.003175534,0.0002871064,0.001068022,0.0003617792],"domain_scores_gemma":[0.99541,0.001175444,0.0007119212,0.0006690244,0.001448114,0.0005854428],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"qualitative","study_design_scores_codex":[0.04032608,0.01282149,0.02616774,0.00191114,0.0002805623,0.00002905671,0.01932163,0.0000193202,0.00004641865,0.002615772,0.0006005145,0.8958603],"study_design_scores_gemma":[0.01435595,0.115058,0.0277872,0.003606794,0.001229674,0.0001716933,0.6252679,0.1545584,0.04673466,0.007007745,0.002726621,0.001495357],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9783038,0.00005675629,0.01544113,0.0007407269,0.001852449,0.003287869,0.00002904837,0.00008495002,0.0002032869],"genre_scores_gemma":[0.9937719,0.00006078222,0.003419667,0.0004276721,0.001744791,0.0004713613,0.00006386369,0.0000174961,0.00002246754],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.894365,"threshold_uncertainty_score":0.8897105,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.404834589605863,"score_gpt":0.5699592721623884,"score_spread":0.1651246825565254,"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."}}