{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002969378,0.0001197695,0.000293314,0.0002803197,0.00005537587,0.00001012388,0.0000760457,0.0000457682,0.00003522572],"category_scores_gemma":[0.000382191,0.00008571429,0.0001889486,0.0006643268,0.00005858161,0.00002352395,0.00002599912,0.0002954931,0.000004263905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009592169,"about_ca_system_score_gemma":0.00002819742,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003809965,"about_ca_topic_score_gemma":0.000002975044,"domain_scores_codex":[0.9989657,0.00003327489,0.0003649668,0.000201503,0.0002257768,0.0002088054],"domain_scores_gemma":[0.9990485,0.0005502697,0.00007328865,0.0001696368,0.00007655408,0.00008174784],"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":[0.00009867225,0.000710553,0.0888439,0.001150295,0.000152641,0.000003282317,0.0003946074,0.4065518,0.001253033,0.000119435,0.0002000208,0.5005217],"study_design_scores_gemma":[0.0007043873,0.0001725372,0.00879745,0.0008303739,0.00006627128,0.000003041871,0.00006602047,0.940037,0.0003742058,0.000002333474,0.04884764,0.0000988068],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7254468,0.002558726,0.267507,0.001574524,0.00188606,0.000782584,0.00001618995,0.0001886709,0.0000394367],"genre_scores_gemma":[0.9977894,0.00015375,0.001763588,0.00002310731,0.0001025534,0.00008652474,0.00005961068,0.00001505699,0.000006379752],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5334851,"threshold_uncertainty_score":0.3495327,"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."}}