{"id":"W4387211695","doi":"10.1007/978-3-031-43898-1_19","title":"Mitigating Calibration Bias Without Fixed Attribute Grouping for Improved Fairness in Medical Imaging Analysis","year":2023,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Cutaneous Melanoma Detection and Management","field":"Medicine","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"NeuroRx Research (Canada); McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Flexibility (engineering); Context (archaeology); Calibration; Artificial intelligence; Trustworthiness; Machine learning; Medical imaging; Software deployment; Cluster (spacecraft); Image (mathematics); Population; Cluster analysis; Baseline (sea); Data mining; Pattern recognition (psychology); Statistics; Medicine; Mathematics; Computer security","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01380152,0.00126506,0.002267519,0.001077053,0.001685311,0.002759724,0.003900799,0.002090631,0.003431596],"category_scores_gemma":[0.04183613,0.0008740182,0.001341355,0.002296034,0.002013067,0.003582726,0.004566723,0.003242556,0.001510234],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001098689,"about_ca_system_score_gemma":0.002185313,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001989919,"about_ca_topic_score_gemma":0.002334228,"domain_scores_codex":[0.9898964,0.005030672,0.000396873,0.002093749,0.001932246,0.0006500593],"domain_scores_gemma":[0.9700274,0.01789292,0.001186626,0.007616758,0.002739563,0.0005366899],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001536355,0.0004954188,0.008407856,0.0003746822,0.0004812129,0.000274839,0.000973338,0.2446616,0.03701764,0.09353641,0.01162972,0.6006109],"study_design_scores_gemma":[0.00004121312,0.0001012483,0.001567171,0.00003630776,0.00009564972,0.000179859,0.00008860083,0.8930206,0.01452047,0.08712278,0.003167558,0.00005858426],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007361282,0.0002969344,0.9912735,0.0001290007,0.00008209812,0.00003040126,0.00005242267,0.0003151627,0.0004591251],"genre_scores_gemma":[0.2712064,0.0004579601,0.7233513,0.0003635034,0.0004534101,0.0001046745,0.0003623469,0.0004869788,0.003213405],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01380152,"threshold_uncertainty_score":0.07299024,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02799182550449574,"score_gpt":0.2850634058637841,"score_spread":0.2570715803592883,"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."}}