{"id":"W2974040794","doi":"10.1515/cdbme-2019-0057","title":"Quantifying the uncertainty of deep learning-based computer-aided diagnosis for patient safety","year":2019,"lang":"en","type":"article","venue":"Current Directions in Biomedical Engineering","topic":"Retinal Imaging and Analysis","field":"Medicine","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"SickKids Foundation; Hospital for Sick Children; University of Toronto","funders":"","keywords":"Artificial intelligence; Computer science; Monte Carlo method; Machine learning; Inference; Pattern recognition (psychology); Statistics; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"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":[0.007607187,0.0007910735,0.0006551376,0.001632595,0.00036418,0.001340995,0.0006898245,0.001324146,0.0008465035],"category_scores_gemma":[0.03997814,0.0003840851,0.0005186605,0.0005091388,0.001124001,0.001604363,0.001455246,0.0009837248,0.0001508524],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001741626,"about_ca_system_score_gemma":0.001181281,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003006498,"about_ca_topic_score_gemma":0.002249845,"domain_scores_codex":[0.9956715,0.001771848,0.0003407354,0.000463672,0.001498806,0.0002533203],"domain_scores_gemma":[0.9519004,0.04073929,0.003380315,0.001233332,0.002343573,0.0004030736],"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":[0.0007919297,0.00009965778,0.03846069,0.0001793212,0.0001549989,0.0002283226,0.0001637559,0.8827634,0.004675337,0.005777395,0.0007723054,0.06593281],"study_design_scores_gemma":[0.000007748937,0.00006685364,0.004388235,0.00003553375,0.00001766893,0.00007246797,0.00002036092,0.9839727,0.004232633,0.006981843,0.0001854575,0.00001851113],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4306273,0.001700163,0.5624839,0.001800454,0.00006913452,0.0000640721,0.0004838303,0.0005389429,0.002232113],"genre_scores_gemma":[0.9887159,0.0001130622,0.01079135,0.0000687631,0.00002215717,0.00001469712,0.0001450096,0.00001268848,0.0001163459],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007607187,"threshold_uncertainty_score":0.04023117,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01906208724024166,"score_gpt":0.2969127915170096,"score_spread":0.2778507042767679,"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."}}