{"id":"W3200676672","doi":"10.1148/ryai.2021210031","title":"A Radiology-focused Review of Predictive Uncertainty for AI Interpretability in Computer-assisted Segmentation","year":2021,"lang":"en","type":"review","venue":"Radiology Artificial Intelligence","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute","funders":"Defence Research and Development Canada","keywords":"Interpretability; Computer science; Artificial intelligence; Segmentation; Machine learning; Deep learning; Data science; Bayesian network; Bayesian probability","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.002278927,0.0005688298,0.004033069,0.0004623802,0.0001008156,0.00001449009,0.0003576853,0.0009288868,0.0003897753],"category_scores_gemma":[0.002696174,0.0005039201,0.0009553534,0.001080406,0.0005912158,0.0001010206,0.00007886034,0.0009791754,0.00003641847],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008853238,"about_ca_system_score_gemma":0.002178925,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002008685,"about_ca_topic_score_gemma":0.0001378099,"domain_scores_codex":[0.9929794,0.001413802,0.003654149,0.001123591,0.000227533,0.0006015159],"domain_scores_gemma":[0.9941357,0.002874514,0.001071446,0.0008140609,0.0009132026,0.0001911053],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"systematic_review","study_design_scores_codex":[0.0001826981,0.0004152874,0.00009014221,0.06824376,0.0002139842,0.00001535613,0.0005272867,0.000062055,0.000005251876,0.0005831542,0.0005742175,0.9290868],"study_design_scores_gemma":[0.0002156004,0.009126374,0.0001270553,0.7213202,0.007736326,0.002513902,0.00199539,0.02558368,0.002163038,0.01087589,0.215556,0.002786532],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.0001225998,0.8978214,0.0946271,0.0007045951,0.001635779,0.004873349,0.0001313576,0.00005047863,0.00003329453],"genre_scores_gemma":[0.0007206765,0.9914795,0.003858359,0.0008279667,0.0006189015,0.001264138,0.001163401,0.00005023924,0.00001682184],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9263003,"threshold_uncertainty_score":0.9997413,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2281720168452039,"score_gpt":0.4874752302087046,"score_spread":0.2593032133635007,"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."}}