{"id":"W4368340908","doi":"10.1016/j.xops.2023.100324","title":"Evaluating the Performance of ChatGPT in Ophthalmology","year":2023,"lang":"en","type":"article","venue":"Ophthalmology Science","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":494,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centre Hospitalier de l’Université de Montréal; Université de Montréal; Centre Intégré Universitaire de Santé et de Services Sociaux du Centre-Sud-de-l'Île-de-Montréal; Hôpital Maisonneuve-Rosemont","funders":"American Academy of Ophthalmology; Bayer","keywords":"Logistic regression; Test (biology); Set (abstract data type); Computer science; Post hoc; Artificial intelligence; Index (typography); Regression; Scale (ratio); Section (typography); Statistics; Test set; Machine learning; Regression analysis; Key (lock); Natural language processing; Medicine; Mathematics; World Wide Web; Programming language","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":[0.0030943,0.00007318005,0.0001705367,0.0003291448,0.0001975074,0.000005908607,0.0002948167,0.00007228829,0.0001885592],"category_scores_gemma":[0.001060828,0.00005183995,0.00002784354,0.002246566,0.001062709,0.0001126466,0.00008389606,0.0002013719,0.0001754863],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006465452,"about_ca_system_score_gemma":0.000630805,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002750812,"about_ca_topic_score_gemma":0.000001414321,"domain_scores_codex":[0.9985737,0.00007864699,0.0003534746,0.0002823662,0.0003036361,0.0004081921],"domain_scores_gemma":[0.9989403,0.0003499508,0.00009786683,0.000354247,0.000194812,0.00006284213],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00009290672,0.00006901012,0.9434685,0.00006046618,0.000002950461,0.00009811488,0.002293526,0.000648941,0.04018906,0.0003490707,0.00002385179,0.01270358],"study_design_scores_gemma":[0.00004940581,0.001493633,0.929229,0.00006733182,0.000006583423,0.003614277,0.00127015,0.0317359,0.03154828,0.0009003478,0.00001158791,0.00007351355],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9932514,0.00005815598,6.792075e-7,0.002275441,0.0006584363,0.0002745027,5.01464e-7,0.00001565205,0.003465208],"genre_scores_gemma":[0.9988925,0.00001682201,0.0002122175,0.00007619089,0.00006402027,0.00004151809,0.000002230158,0.000005471554,0.0006889753],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03108696,"threshold_uncertainty_score":0.3915595,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3596977566977329,"score_gpt":0.5413077638481226,"score_spread":0.1816100071503898,"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."}}