{"id":"W4200501894","doi":"10.21203/rs.3.rs-943804/v1","title":"Biased Data, Biased AI: Deep Networks Predict the Acquisition Site of TCGA Images","year":2021,"lang":"en","type":"preprint","venue":"Research Square","topic":"AI in cancer detection","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Tech University; McMaster University; University of Waterloo","funders":"","keywords":"Digital pathology; Computer science; Demographics; Artificial intelligence; Deep learning; Scope (computer science); Machine learning; Data science; Pattern recognition (psychology); Demography","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.00317984,0.0006421647,0.0004076506,0.0007795087,0.0003263091,0.001333641,0.0009223762,0.001345497,0.001296968],"category_scores_gemma":[0.01948471,0.0003176175,0.0003462204,0.0006959146,0.0008516224,0.001377704,0.001185257,0.001786003,0.0004226632],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001062544,"about_ca_system_score_gemma":0.0007615028,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005350042,"about_ca_topic_score_gemma":0.004802969,"domain_scores_codex":[0.9990758,0.0003207737,0.00004718066,0.0002607971,0.0001695845,0.0001258409],"domain_scores_gemma":[0.9940683,0.003625718,0.0006048062,0.0006679623,0.0008191135,0.0002141584],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001640717,0.0002816444,0.1718503,0.0003200138,0.0003136553,0.0006326332,0.0003017272,0.5530335,0.01723292,0.01385762,0.01749277,0.2230425],"study_design_scores_gemma":[0.00002053589,0.00005437488,0.008029449,0.00002908746,0.00002832054,0.0001002793,0.00005146275,0.9752067,0.005174622,0.0104202,0.0008704745,0.00001446096],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8027572,0.002071262,0.1846371,0.003241759,0.0003597155,0.00006344586,0.001337652,0.00108615,0.004445625],"genre_scores_gemma":[0.9876852,0.000150492,0.009813896,0.0002432681,0.00006006178,0.00001978854,0.0007913817,0.00004208199,0.001193743],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9968202,"threshold_uncertainty_score":0.01681679,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07956605645229957,"score_gpt":0.3862819758299171,"score_spread":0.3067159193776175,"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."}}