{"id":"W3107625569","doi":"10.1016/j.media.2020.101912","title":"Learning to segment images with classification labels","year":2020,"lang":"en","type":"article","venue":"Medical Image Analysis","topic":"AI in cancer detection","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":false,"ca_institutions":"Sunnybrook Health Science Centre; University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Cancer Society","keywords":"Computer science; Segmentation; Artificial intelligence; Ground truth; Annotation; Class (philosophy); Task (project management); Pattern recognition (psychology); Image segmentation; Labeled data; Machine learning","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":[0.0003302044,0.0001048773,0.0002062513,0.0001505332,0.0001036131,0.0001490552,0.000563387,0.00004493176,0.0003000562],"category_scores_gemma":[0.000297784,0.00008383366,0.00008085748,0.002840446,0.00004916187,0.0003058654,0.0001686885,0.000234081,0.0002147427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006918192,"about_ca_system_score_gemma":0.00006813659,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004213213,"about_ca_topic_score_gemma":0.00001920409,"domain_scores_codex":[0.9981416,0.0001104781,0.0002002855,0.0004772985,0.0008706614,0.0001996268],"domain_scores_gemma":[0.9990355,0.00006473825,0.00007692337,0.0003055826,0.0001182527,0.0003990461],"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":[0.00006629092,0.0001624974,0.02371165,0.00006184959,0.001796466,0.0002909957,0.006902458,0.007507836,0.02646209,0.0003918222,0.01385945,0.9187866],"study_design_scores_gemma":[0.0003993491,0.0004851446,0.01603515,0.00002076317,0.0004307128,0.00000578009,0.0003676933,0.9598336,0.01218468,0.00003900066,0.009863051,0.0003351215],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008840118,0.00002329205,0.9540822,0.03596611,0.00002765306,0.00007141878,6.296569e-7,0.0002100839,0.0007784931],"genre_scores_gemma":[0.9284187,0.00002207787,0.06730404,0.00384834,0.0001371815,0.00003339992,0.000006288426,0.000009626745,0.0002203574],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9523257,"threshold_uncertainty_score":0.3418637,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01534197667013368,"score_gpt":0.2692718852778553,"score_spread":0.2539299086077216,"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."}}