{"id":"W4205679507","doi":"10.1007/s10140-022-02019-3","title":"Deep learning prediction of sex on chest radiographs: a potential contributor to biased algorithms","year":2022,"lang":"en","type":"article","venue":"Emergency Radiology","topic":"COVID-19 diagnosis using AI","field":"Medicine","cited_by":17,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa","funders":"Johns Hopkins University","keywords":"Convolutional neural network; Artificial intelligence; Receiver operating characteristic; Medicine; Test set; Undersampling; Machine learning; Deep learning; Radiography; Set (abstract data type); Pattern recognition (psychology); F1 score; Computer science; Radiology","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":[0.01398951,0.001008743,0.001148392,0.0008661382,0.0007778013,0.001996124,0.002434658,0.002272401,0.00263678],"category_scores_gemma":[0.08356715,0.0005616404,0.000580337,0.0008512374,0.0008008128,0.001530653,0.001269389,0.002778667,0.001630115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001011847,"about_ca_system_score_gemma":0.002056692,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009520772,"about_ca_topic_score_gemma":0.01192437,"domain_scores_codex":[0.9937556,0.003330758,0.0004586044,0.001165288,0.0009041371,0.0003856178],"domain_scores_gemma":[0.9590234,0.03117678,0.001488577,0.003500637,0.00432271,0.0004879405],"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":[0.0020085,0.0007571903,0.2398355,0.0007473047,0.001013527,0.0006859979,0.0004369495,0.1508365,0.007918855,0.02122369,0.03848775,0.5360482],"study_design_scores_gemma":[0.0001647288,0.0002338647,0.02966525,0.0003404191,0.0002447405,0.0008138784,0.0001344316,0.9115151,0.009075869,0.04041401,0.007343399,0.00005426526],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4113707,0.0120206,0.5318119,0.01848291,0.001718039,0.0005320033,0.004371037,0.003749733,0.01594315],"genre_scores_gemma":[0.9349487,0.0007537493,0.05424374,0.003420104,0.0004095267,0.0001537944,0.002056799,0.0002206565,0.003792934],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01398951,"threshold_uncertainty_score":0.0739845,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02010145942009194,"score_gpt":0.2892124169157456,"score_spread":0.2691109574956536,"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."}}