{"id":"W4320481778","doi":"10.1007/978-3-031-25069-9_13","title":"FairDisCo: Fairer AI in Dermatology via Disentanglement Contrastive Learning","year":2023,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Cutaneous Melanoma Detection and Management","field":"Medicine","cited_by":28,"is_retracted":false,"has_abstract":false,"ca_institutions":"Simon Fraser University; University of British Columbia","funders":"","keywords":"Skin lesion; Computer science; Resampling; Artificial intelligence; Deep learning; Lesion; Feature (linguistics); Pattern recognition (psychology); Machine learning; Task (project management); Dermatology; Medicine; Pathology","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0002432953,0.0003324226,0.0005442432,0.0009025881,0.000108771,0.00006520236,0.0002608171,0.0001933669,0.00009123033],"category_scores_gemma":[0.00006705349,0.0003010352,0.0001032675,0.0003457131,0.0003619439,0.00006776224,0.0003929191,0.0009105716,0.0001349105],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004290967,"about_ca_system_score_gemma":0.0001519293,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004718959,"about_ca_topic_score_gemma":0.0006203422,"domain_scores_codex":[0.9977143,0.00002908464,0.0004222197,0.0008199685,0.0005277581,0.0004867066],"domain_scores_gemma":[0.9991423,0.0001924191,0.0001299794,0.0003281469,0.00007998569,0.0001271118],"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":[0.0002099404,0.0001364141,0.003842829,0.0003061695,0.0001281926,0.01014857,0.001948899,0.01891297,0.000968216,0.003739177,0.0001962043,0.9594624],"study_design_scores_gemma":[0.008856694,0.002901067,0.01894132,0.005317884,0.0003935351,0.007140676,0.00004485814,0.7956827,0.005105498,0.0963369,0.05545865,0.003820204],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001203669,0.0001204437,0.9849087,0.004431217,0.001896602,0.0009769902,0.000002707267,0.0002095693,0.006250078],"genre_scores_gemma":[0.9906928,0.00004583798,0.002625912,0.002497368,0.0002762314,0.00002401574,0.00001992274,0.00005467276,0.003763223],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9894891,"threshold_uncertainty_score":0.9999442,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01213528635741349,"score_gpt":0.258569938422091,"score_spread":0.2464346520646775,"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."}}