{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005034073,0.0009925247,0.001591581,0.003953105,0.000383033,0.002129833,0.001429581,0.002045668,0.005111008],"category_scores_gemma":[0.02016747,0.0005502892,0.001177485,0.003594252,0.001559795,0.002383747,0.001053571,0.00201617,0.001337642],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001879042,"about_ca_system_score_gemma":0.003621355,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003246951,"about_ca_topic_score_gemma":0.003952418,"domain_scores_codex":[0.9982672,0.0007327521,0.0002924203,0.0002072983,0.0004504894,0.00004985628],"domain_scores_gemma":[0.9794862,0.0180221,0.0007666983,0.0002056088,0.001393257,0.0001261517],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004965374,0.00003517685,0.000382829,0.0506728,0.0002673184,0.0001136724,0.0001801804,0.001451943,0.0001853021,0.01978585,0.02408634,0.9027889],"study_design_scores_gemma":[0.00002731751,0.0001461936,0.002267925,0.1050101,0.0008478194,0.001636027,0.000232555,0.00179254,0.0006282045,0.03332558,0.8539709,0.0001148532],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00008110435,0.9964122,0.001065202,0.001083863,0.0001762253,0.000009662139,0.00002433541,0.00000871161,0.001138762],"genre_scores_gemma":[0.002166784,0.9957317,0.001000041,0.000509134,0.0003235728,0.00002023721,0.00003279893,0.000007667719,0.0002081224],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.005111008,"threshold_uncertainty_score":0.02662307,"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."}}