{"id":"W2799738340","doi":"10.1016/j.media.2019.02.009","title":"Constrained-CNN losses for weakly supervised segmentation","year":2019,"lang":"en","type":"article","venue":"Medical Image Analysis","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":282,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo; Kahnawake Education Center","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Leverage (statistics); Computer science; Segmentation; Iterated function; Mathematical optimization; Differentiable function; Artificial intelligence; Dual (grammatical number); Pixel; Machine learning; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001638822,0.001655426,0.001566969,0.000909861,0.0005054626,0.001427254,0.002458518,0.003053137,0.007311325],"category_scores_gemma":[0.004553595,0.000944889,0.0009362351,0.0009504641,0.0009293265,0.001999763,0.002223662,0.002169422,0.002170645],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001502809,"about_ca_system_score_gemma":0.001970855,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0110363,"about_ca_topic_score_gemma":0.01649029,"domain_scores_codex":[0.9994029,0.0001312025,0.00003305747,0.0001872794,0.0001514737,0.00009425543],"domain_scores_gemma":[0.9988948,0.0004736722,0.0000837022,0.0002218005,0.0002483404,0.00007770323],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00070238,0.0002389683,0.001191642,0.0003721795,0.0001807636,0.0002141122,0.00008170997,0.4723098,0.01568294,0.02289036,0.02251445,0.4636207],"study_design_scores_gemma":[0.000008412842,0.00002167177,0.0001916292,0.00001690859,0.00001005243,0.00003577516,0.000005969469,0.9905709,0.002089133,0.006043135,0.001000841,0.000005471987],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02602418,0.001257261,0.96405,0.0006452005,0.0001363556,0.0001341866,0.0009780059,0.002471222,0.004303598],"genre_scores_gemma":[0.565331,0.001353179,0.3839627,0.0009307563,0.0003267086,0.0004426646,0.00639524,0.001552613,0.0397051],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0110363,"threshold_uncertainty_score":0.02445877,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0113012703349794,"score_gpt":0.2936607132017627,"score_spread":0.2823594428667833,"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."}}