{"id":"W4391335149","doi":"10.48550/arxiv.2401.14497","title":"Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasets","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Cutaneous Melanoma Detection and Management","field":"Medicine","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Quality (philosophy); Computer science; Image (mathematics); Artificial intelligence; Dermatology; Medicine; Physics","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":[0.007987889,0.001046076,0.0008885285,0.00519282,0.001252095,0.002597792,0.002356352,0.001805043,0.001903894],"category_scores_gemma":[0.03051507,0.0003522844,0.001207052,0.003715276,0.001453958,0.002010186,0.002608806,0.001807626,0.001328851],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002081202,"about_ca_system_score_gemma":0.001551628,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01621926,"about_ca_topic_score_gemma":0.02311705,"domain_scores_codex":[0.9916276,0.0016057,0.0008829053,0.001823032,0.003538011,0.0005227515],"domain_scores_gemma":[0.9849124,0.005594147,0.001075705,0.003386547,0.004364547,0.0006666942],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.004208616,0.001354564,0.1696617,0.003925607,0.00230779,0.002255243,0.001158568,0.05634331,0.04199648,0.006822511,0.4117002,0.2982655],"study_design_scores_gemma":[0.0007365944,0.001083366,0.3644889,0.00123543,0.0007135502,0.008126796,0.003668821,0.2842954,0.1132431,0.01442577,0.2075042,0.0004780613],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7974658,0.01047145,0.02993857,0.008318521,0.001247884,0.0005560633,0.1262556,0.01194795,0.01379807],"genre_scores_gemma":[0.6526283,0.001121479,0.05041842,0.001335788,0.0001519798,0.0002968552,0.2890838,0.001480845,0.003482596],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9920121,"threshold_uncertainty_score":0.04224449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1545767793963772,"score_gpt":0.260344629495632,"score_spread":0.1057678500992548,"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."}}