{"id":"W3105141008","doi":"10.1109/tmi.2021.3123461","title":"Deep Interpretable Classification and Weakly-Supervised Segmentation of Histology Images via Max-Min Uncertainty","year":2021,"lang":"en","type":"article","venue":"IEEE Transactions on Medical Imaging","topic":"AI in cancer detection","field":"Computer Science","cited_by":67,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Compute Canada","keywords":"Discriminative model; Artificial intelligence; Pattern recognition (psychology); Computer science; Segmentation; Kullback–Leibler divergence; Support vector machine; Robustness (evolution); Contextual image classification; Machine learning; 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":[0.0002856539,0.0001310178,0.0001937996,0.0001537354,0.0001398585,0.00004727464,0.0002551108,0.00008040859,0.0002085548],"category_scores_gemma":[0.00003008764,0.0001370699,0.00006321687,0.0003805567,0.0001951833,0.0003733611,0.000007298294,0.0002999558,0.0000106628],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001852424,"about_ca_system_score_gemma":0.0001485619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001053135,"about_ca_topic_score_gemma":0.00007184078,"domain_scores_codex":[0.9984123,0.000178168,0.0003480785,0.0004340791,0.0004228759,0.0002044888],"domain_scores_gemma":[0.9990726,0.0002073685,0.0000979285,0.0003434195,0.0001464813,0.0001322786],"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":[0.0000187043,0.0001161589,0.00009732353,0.00005286008,0.00002963041,0.00002037766,0.0007349251,0.001283619,0.1656516,0.0001042049,0.0000793435,0.8318112],"study_design_scores_gemma":[0.0004786593,0.00004042529,0.0003163544,0.0000674816,0.00002984831,0.0001532393,0.0003171218,0.8561912,0.1415528,0.0005285185,0.0001870678,0.0001373235],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006358543,0.0002818609,0.9882078,0.003681531,0.001015588,0.000105311,0.000002524915,0.00008750203,0.0002593297],"genre_scores_gemma":[0.9747064,0.0001796655,0.02441918,0.0005311224,0.00002966533,0.00004107879,0.000002698916,0.00001149918,0.00007867102],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9683478,"threshold_uncertainty_score":0.5589548,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01160283661625003,"score_gpt":0.2640108282827661,"score_spread":0.2524079916665161,"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."}}