{"id":"W3044795413","doi":"10.1016/j.compmedimag.2020.101765","title":"Automatic skin lesion classification based on mid-level feature learning","year":2020,"lang":"en","type":"article","venue":"Computerized Medical Imaging and Graphics","topic":"Cutaneous Melanoma Detection and Management","field":"Medicine","cited_by":73,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Artificial intelligence; Discriminative model; Computer science; Pattern recognition (psychology); Convolutional neural network; Feature (linguistics); Segmentation; Deep learning; Similarity (geometry); Metric (unit); Class (philosophy); Skin lesion; Feature learning; Feature extraction; Image (mathematics)","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0002179152,0.0001533619,0.0002478714,0.0001402346,0.0001429166,0.00005001353,0.00007685352,0.00009278656,0.00007581523],"category_scores_gemma":[0.0002870381,0.0001286077,0.00008250309,0.000275211,0.0001029393,0.00002746075,0.00005094516,0.0005348435,0.00001423459],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001928772,"about_ca_system_score_gemma":0.00005325511,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003971114,"about_ca_topic_score_gemma":5.165908e-7,"domain_scores_codex":[0.99872,0.00009080447,0.0001936939,0.0003250144,0.0004934854,0.0001770019],"domain_scores_gemma":[0.999186,0.0001220671,0.00007012529,0.0001457273,0.00004741448,0.0004286738],"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":[0.000318246,0.0002249908,0.002897623,0.0006425824,0.00006233009,0.0005515103,0.0003977882,0.00002757334,0.004098647,0.0005717574,0.01544042,0.9747666],"study_design_scores_gemma":[0.002114236,0.0002040642,0.0157282,0.0003007759,0.00005477338,0.0001122557,0.00006854357,0.9396278,0.000128645,0.0000325703,0.04149539,0.0001327031],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2693156,0.0004094183,0.4513331,0.2732665,0.001083935,0.0009363022,0.000005521027,0.001460619,0.002189005],"genre_scores_gemma":[0.9770287,0.000100923,0.003766899,0.01867135,0.0002521078,0.000008441479,0.00004603836,0.00002151,0.0001040461],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9746338,"threshold_uncertainty_score":0.5244469,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03587550213371445,"score_gpt":0.2818451848954405,"score_spread":0.245969682761726,"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."}}