{"id":"W4313643958","doi":"10.1007/s00138-022-01369-9","title":"Pixel-wise confidence estimation for segmentation in Bayesian Convolutional Neural Networks","year":2023,"lang":"en","type":"article","venue":"Machine Vision and Applications","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Uncertainty quantification; Pixel; Overfitting; Computer science; Weighting; Artificial intelligence; Convolutional neural network; Bayesian probability; Confidence interval; Thresholding; Robust confidence intervals; Sensitivity analysis; Entropy (arrow of time); Measurement uncertainty; Uncertainty analysis; Pattern recognition (psychology); Machine learning; Artificial neural network; Statistics; Mathematics; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005891697,0.001144061,0.002340774,0.002042464,0.0007197083,0.002435678,0.00355097,0.003253966,0.002839539],"category_scores_gemma":[0.03403697,0.002091446,0.001184154,0.001466345,0.001880631,0.003440488,0.002637449,0.003321919,0.0006998309],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002706629,"about_ca_system_score_gemma":0.002378761,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01185464,"about_ca_topic_score_gemma":0.01219949,"domain_scores_codex":[0.9982754,0.0004969693,0.0001274606,0.0004473551,0.0004639939,0.0001888214],"domain_scores_gemma":[0.9851136,0.01153496,0.001004825,0.0006810377,0.001342818,0.0003227236],"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":[0.0005116633,0.00007436763,0.002235288,0.0002465092,0.0001420121,0.00007236114,0.0001247575,0.8220953,0.004428033,0.03246833,0.001728178,0.1358732],"study_design_scores_gemma":[0.000006712173,0.000008180818,0.0001994601,0.0000176962,0.000007791331,0.00001209522,0.000003054069,0.9902385,0.000899031,0.00846571,0.0001345048,0.000007207356],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01239565,0.0004438031,0.9858302,0.0002031511,0.00002440012,0.00002604776,0.0001023823,0.0004606189,0.0005137661],"genre_scores_gemma":[0.6078528,0.0008100294,0.3847888,0.0002981929,0.0001899738,0.0002118547,0.00129194,0.0006282683,0.003928076],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01185464,"threshold_uncertainty_score":0.03115863,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01393376341664312,"score_gpt":0.3162321342349305,"score_spread":0.3022983708182874,"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."}}