{"id":"W4401540052","doi":"10.1073/pnas.2402267121","title":"Race adjustments in clinical algorithms can help correct for racial disparities in data quality","year":2024,"lang":"en","type":"article","venue":"Proceedings of the National Academy of Sciences","topic":"Colorectal Cancer Screening and Detection","field":"Medicine","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Cancer Institute; Canadian Institute for Advanced Research; Google","keywords":"Race (biology); Quality (philosophy); Computer science; Algorithm; Sociology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"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":[0.003992805,0.00006190259,0.0002006808,0.0001785539,0.00005117269,0.00001948515,0.0003480504,0.00007994463,0.000003311767],"category_scores_gemma":[0.002124938,0.0000429565,0.00005769593,0.000711412,0.0003508881,0.0002748707,0.0001298172,0.0002336883,2.236835e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007505128,"about_ca_system_score_gemma":0.0001175266,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001699874,"about_ca_topic_score_gemma":0.00001992647,"domain_scores_codex":[0.9984657,0.00001125659,0.0004545669,0.0002970075,0.0006524869,0.0001190514],"domain_scores_gemma":[0.9993371,0.0003774569,0.0001436375,0.00001108584,0.0001035874,0.00002711453],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.005332625,0.000724211,0.775394,0.0023617,0.0001361726,2.381432e-7,0.002305954,0.000142562,0.01855752,0.02332968,0.01505058,0.1566648],"study_design_scores_gemma":[0.0008963946,0.0004406976,0.93369,0.0007066454,0.00002885904,0.00001007644,0.0004450747,0.03761949,0.01322313,0.0119506,0.0008848942,0.000104198],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9903487,0.0005920231,0.00000981919,0.007735405,0.0001609952,0.0003693219,0.00008374677,0.00001536849,0.0006846015],"genre_scores_gemma":[0.9983122,0.00005895805,0.001017347,0.0001923538,0.0001939262,0.00002287108,0.000001369282,0.000003013525,0.0001979567],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1582959,"threshold_uncertainty_score":0.2543902,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2098348985683694,"score_gpt":0.4691904542167525,"score_spread":0.2593555556483831,"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."}}