{"id":"W4400163580","doi":"10.1093/jamia/ocae165","title":"Towards objective and systematic evaluation of bias in artificial intelligence for medical imaging","year":2024,"lang":"en","type":"article","venue":"Journal of the American Medical Informatics Association","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hotchkiss Brain Institute; Alberta Children's Hospital; University of Calgary","funders":"Alberta Innovates; Alberta Children's Hospital Foundation; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Children's Hospital Foundation","keywords":"Artificial intelligence; Computer science; Medical imaging; Machine learning; Data science","routes":{"ca_aff":true,"ca_fund":true,"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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01608148,0.00008359257,0.0004575883,0.0002720494,0.00004213366,0.00003769569,0.0001371552,0.00008498832,0.00002607844],"category_scores_gemma":[0.03974567,0.00005250997,0.0001375548,0.0005770237,0.0001258127,0.0001860529,0.00002861229,0.0004206124,0.000002262684],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007818706,"about_ca_system_score_gemma":0.002154324,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001515974,"about_ca_topic_score_gemma":0.00005731254,"domain_scores_codex":[0.9952592,0.0003028413,0.001878775,0.00005779057,0.002341743,0.0001596405],"domain_scores_gemma":[0.9952797,0.00200475,0.001392623,0.0001006908,0.001071276,0.0001510117],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001399089,0.0001837504,0.01557933,0.007362116,0.000232142,0.000004543824,0.02172424,0.0001520907,0.00004509231,0.001636281,0.0006420174,0.9522985],"study_design_scores_gemma":[0.0001026388,0.0003030498,0.005262423,0.01238268,0.0005239749,0.0001552862,0.02906488,0.9333428,0.001776463,0.01695057,0.00003734148,0.00009790514],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9598497,0.0007521185,0.021359,0.01594703,0.001176724,0.0007618317,0.000003409752,0.00000876958,0.0001414228],"genre_scores_gemma":[0.9981807,0.0002637557,0.0006224418,0.0006155575,0.0002840642,0.00001874124,0.00000175492,0.000007127709,0.000005885323],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9522006,"threshold_uncertainty_score":0.968343,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1236259313214339,"score_gpt":0.4627124433982143,"score_spread":0.3390865120767803,"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."}}