{"id":"W4309838302","doi":"10.1038/s41598-022-24574-y","title":"Quantifying deep neural network uncertainty for atrial fibrillation detection with limited labels","year":2022,"lang":"en","type":"article","venue":"Scientific Reports","topic":"ECG Monitoring and Analysis","field":"Medicine","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; Queen's University","funders":"Southeastern Ontario Academic Medical Organization; Natural Sciences and Engineering Research Council of Canada; Queen's University","keywords":"Atrial fibrillation; Computer science; Artificial intelligence; Artificial neural network; Cardiology; Internal medicine; Medicine; Pattern recognition (psychology)","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":[],"consensus_categories":[],"category_scores_codex":[0.001370251,0.0001044247,0.0002128627,0.0001575274,0.001153722,0.0001114978,0.00003698584,0.00003378942,0.00004613136],"category_scores_gemma":[0.0001759707,0.00008773582,0.0002001996,0.001028462,0.00005308364,0.00007504871,0.00003769653,0.0001308825,0.000001126285],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001209184,"about_ca_system_score_gemma":0.00007513701,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006262431,"about_ca_topic_score_gemma":0.00008206427,"domain_scores_codex":[0.9982737,0.00005876962,0.0003458578,0.0005272025,0.0005163979,0.0002780668],"domain_scores_gemma":[0.9989325,0.00007285715,0.000291649,0.0004174839,0.0001962557,0.00008919818],"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":[0.0005361367,0.00002121496,0.1376507,0.00005099228,0.0001557311,0.0001721076,0.0002377007,0.7935649,0.01428739,0.000005061476,0.0007106272,0.05260735],"study_design_scores_gemma":[0.001856424,0.001064539,0.007711426,0.00008457556,0.00106661,0.00136151,0.0007823021,0.8963917,0.006415057,0.000690481,0.08208574,0.0004895943],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9894115,0.0002749072,0.002571891,0.000148105,0.006985951,0.0004394428,0.000001567419,0.0001171919,0.00004942693],"genre_scores_gemma":[0.9964601,0.000001333925,0.001011178,0.00001424241,0.001246983,0.00002352016,0.0001590297,0.00001791401,0.001065725],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1299393,"threshold_uncertainty_score":0.8873616,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03938516488153625,"score_gpt":0.2956681694548877,"score_spread":0.2562830045733514,"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."}}