{"id":"W4220692674","doi":"10.1503/cmaj.80860","title":"Biased data lead to biased algorithms","year":2022,"lang":"en","type":"letter","venue":"Canadian Medical Association Journal","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Lead (geology); Data science; Machine learning; Subject (documents); Artificial intelligence; Series (stratigraphy); Algorithm; World Wide Web","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.03877651,0.0007298291,0.001844452,0.00183725,0.002409743,0.006967226,0.00213067,0.0202334,0.006544341],"category_scores_gemma":[0.3477095,0.0007257945,0.001075401,0.001440812,0.01241467,0.007631087,0.002230765,0.03138721,0.004954982],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003603797,"about_ca_system_score_gemma":0.0029221,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003371719,"about_ca_topic_score_gemma":0.003425664,"domain_scores_codex":[0.9633427,0.02180478,0.002168322,0.003203147,0.008239238,0.001241806],"domain_scores_gemma":[0.682025,0.2729754,0.006909666,0.01277788,0.02209326,0.003218836],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0004388221,0.0001025179,0.006848179,0.0003334607,0.000240067,0.001474358,0.001169109,0.002526656,0.0003426303,0.3222887,0.5533381,0.1108974],"study_design_scores_gemma":[0.0003513164,0.00005947157,0.001222478,0.000604933,0.0000651138,0.001813497,0.0004609807,0.006697724,0.0005924545,0.8086307,0.1794203,0.00008104093],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"commentary","genre_gemma":"commentary","genre_scores_codex":[0.001705853,0.002700392,0.01057102,0.9703615,0.005232029,0.00002770323,0.0001059327,0.00006272112,0.009232932],"genre_scores_gemma":[0.07197079,0.004351946,0.01840436,0.8652642,0.03342727,0.0002024739,0.0001475312,0.0002609181,0.005970433],"genre_candidate":"commentary","genre_consensus":"commentary","teacher_disagreement_score":0.9612235,"threshold_uncertainty_score":0.2050722,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.205440489926531,"score_gpt":0.4090220843540529,"score_spread":0.2035815944275219,"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."}}