{"id":"W4401173939","doi":"10.3390/bioengineering11080770","title":"An Unsupervised Error Detection Methodology for Detecting Mislabels in Healthcare Analytics","year":2024,"lang":"en","type":"article","venue":"Bioengineering","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Interpretability; Computer science; Cluster analysis; Unsupervised learning; Artificial intelligence; Data mining; Reliability (semiconductor); Machine learning; Pattern recognition (psychology)","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":[0.001989961,0.0002025293,0.0003278349,0.0004996786,0.0003372075,0.00002046628,0.0001761803,0.0004009038,0.00003195393],"category_scores_gemma":[0.0008721071,0.0002097688,0.00007892154,0.0007172154,0.00002239973,0.0001907464,0.00003483232,0.0008439381,0.00004388965],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000538854,"about_ca_system_score_gemma":0.0002211381,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003153406,"about_ca_topic_score_gemma":0.01098571,"domain_scores_codex":[0.9973412,0.0004541205,0.000791424,0.0004717426,0.0001307381,0.0008107956],"domain_scores_gemma":[0.9974796,0.001793298,0.00006365569,0.0003154657,0.000164943,0.0001830622],"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":[0.0003707687,0.00009302326,0.0472542,0.01723204,0.000129738,0.00009542114,0.03416065,0.03531289,0.3394498,0.01251832,0.0001046851,0.5132784],"study_design_scores_gemma":[0.0001497843,0.000289555,0.001558662,0.0006523456,0.00002095189,0.000005323509,0.005893082,0.9712765,0.01549276,0.001788767,0.002560387,0.000311855],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7277863,0.001529751,0.2626795,0.002028909,0.003730025,0.00146671,0.0000364116,0.0007185466,0.00002380329],"genre_scores_gemma":[0.9835122,0.00003921692,0.01496488,0.000239121,0.0006935512,0.0004062253,0.00001317485,0.00008335144,0.00004828343],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9359636,"threshold_uncertainty_score":0.8554122,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3677459445365586,"score_gpt":0.5273353419595005,"score_spread":0.1595893974229419,"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."}}