{"id":"W3122802246","doi":"","title":"Machine learning to identify multifocal ERG deficits in patients taking hydroxychloroquine.","year":2019,"lang":"en","type":"article","venue":"Investigative Ophthalmology & Visual Science","topic":"Drug-Induced Ocular Toxicity","field":"Medicine","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Hydroxychloroquine; Erg; Medicine; Ophthalmology; Optometry; Computer science; Internal medicine; Coronavirus disease 2019 (COVID-19); Retinal","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001372502,0.0003482289,0.0005689587,0.0006971636,0.0002303129,0.00005505754,0.0005140824,0.0001876082,0.0001878399],"category_scores_gemma":[0.003799389,0.0003264092,0.00008528057,0.002227535,0.001200809,0.0004695599,0.0005323865,0.0008555481,0.0009315998],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004997865,"about_ca_system_score_gemma":0.0003534062,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004540903,"about_ca_topic_score_gemma":0.00001606705,"domain_scores_codex":[0.996112,0.0002667004,0.0005322926,0.001163161,0.0009662302,0.0009595988],"domain_scores_gemma":[0.9982391,0.0001654874,0.0002770428,0.0003822546,0.0003122939,0.0006238394],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00004763969,0.0001855949,0.6013505,0.0000153753,0.000007985195,0.0001159289,0.0007584964,0.0002498513,0.3967457,0.00002505286,0.000001534702,0.0004963547],"study_design_scores_gemma":[0.001331834,0.001142681,0.8588941,0.0001908153,0.00001301953,0.00009620237,0.0001480944,0.003804764,0.1338743,0.0001827766,0.00002824803,0.0002932038],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9942835,0.0000280964,0.000003755358,0.0002641756,0.0004894054,0.0009807822,0.000004503262,0.00006923865,0.003876574],"genre_scores_gemma":[0.9978145,5.364341e-7,0.001164924,0.0005456757,0.00004471462,0.00003094658,0.0000177818,0.00003773245,0.0003431639],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2628714,"threshold_uncertainty_score":0.9999188,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03977036196271036,"score_gpt":0.3631294103394533,"score_spread":0.3233590483767429,"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."}}