{"id":"W4410850908","doi":"10.1177/0272989x251343082","title":"So You’ve Got a High AUC, Now What? An Overview of Important Considerations when Bringing Machine-Learning Models from Computer to Bedside","year":2025,"lang":"en","type":"review","venue":"Medical Decision Making","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary; Public Health Ontario; University of Toronto","funders":"","keywords":"Machine learning; Computer science; Artificial intelligence; Bridge (graph theory); Software deployment; Key (lock); Model selection; Clinical decision support system; Health care; Clinical Practice; Calibration; Medical physics; Decision support system; Medicine; Software engineering","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001692638,0.000483527,0.0023405,0.0006484721,0.000313381,0.0002386682,0.0003577743,0.0006376266,0.001889294],"category_scores_gemma":[0.003547473,0.0003978835,0.0004560048,0.0005142626,0.00009114331,0.0003213011,0.0003225559,0.001259387,0.00007527945],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002711796,"about_ca_system_score_gemma":0.002397014,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001161979,"about_ca_topic_score_gemma":0.0007495603,"domain_scores_codex":[0.994032,0.0004231693,0.002774841,0.0009070921,0.001375807,0.0004871478],"domain_scores_gemma":[0.9927918,0.004697111,0.0006690953,0.0008025854,0.0004146705,0.0006247707],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"systematic_review","study_design_scores_codex":[0.00002409768,0.0001043743,0.00004456714,0.003344257,0.00009079561,0.0001202698,0.001217939,0.0001595852,3.369556e-7,0.0007218627,0.001214205,0.9929577],"study_design_scores_gemma":[0.0001933781,0.0005272419,0.00002950552,0.4769529,0.001543252,0.0002163826,0.0006254354,0.119282,0.00001045614,0.04930885,0.3504392,0.0008714826],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.000353214,0.888135,0.1073178,0.00106845,0.001898701,0.001049568,0.00004109086,0.00008879648,0.00004740678],"genre_scores_gemma":[0.001565513,0.9538375,0.04047652,0.002506384,0.001063389,0.00008104817,0.0003174763,0.00006655893,0.00008568388],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9920862,"threshold_uncertainty_score":0.9998473,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3326424958883833,"score_gpt":0.5007955916939023,"score_spread":0.168153095805519,"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."}}