{"id":"W4388879751","doi":"10.1097/iop.0000000000002552","title":"Evaluating ChatGPT on Orbital and Oculofacial Disorders: Accuracy and Readability Insights","year":2023,"lang":"en","type":"article","venue":"Ophthalmic Plastic and Reconstructive Surgery","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; Ottawa Hospital; Queen's University; University of Ottawa; Dalhousie University; University of Toronto","funders":"","keywords":"Readability; Medicine; Intraclass correlation; Reliability (semiconductor); Likert scale; Artificial intelligence; Statistics; Medical physics; Natural language processing; Psychometrics; Computer science; Clinical 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":[0.0004125792,0.0001636052,0.0003319007,0.0002239489,0.0002701208,0.00003243101,0.00001976043,0.0001222031,0.00004350899],"category_scores_gemma":[0.003741252,0.0001341848,0.000042023,0.0002380942,0.0003950319,0.0001558688,0.00003646116,0.0002118372,0.00002141493],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003902048,"about_ca_system_score_gemma":0.0001930417,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001713028,"about_ca_topic_score_gemma":0.0000157615,"domain_scores_codex":[0.9987066,0.000089568,0.0003635348,0.000424833,0.0001651922,0.0002502705],"domain_scores_gemma":[0.9946782,0.00482549,0.0001080156,0.0001207933,0.00008236191,0.000185121],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0003766888,0.00005619821,0.4410231,0.0002325409,0.00004430651,0.00003221215,0.001941624,0.000003071176,0.0004975127,0.0001784068,0.00008142358,0.5555329],"study_design_scores_gemma":[0.0001673097,0.000654799,0.9626135,0.0007111236,0.00009598803,0.001289224,0.01082466,0.004270065,0.001407876,0.0173937,0.0001376873,0.0004340545],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9974532,0.0002740088,0.0000107537,0.000378241,0.001013667,0.0002887495,0.00001318395,0.00004755928,0.0005206087],"genre_scores_gemma":[0.9990005,0.0005527175,0.00009221812,0.00003222663,0.0002042296,0.00002676076,0.00002529987,0.00001328045,0.00005280643],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5550989,"threshold_uncertainty_score":0.5471897,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1639992599863989,"score_gpt":0.4172744711122012,"score_spread":0.2532752111258023,"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."}}