{"id":"W4394760737","doi":"10.1001/jamaophthalmol.2024.0017","title":"Accuracy of an Artificial Intelligence Chatbot’s Interpretation of Clinical Ophthalmic Images","year":2024,"lang":"en","type":"article","venue":"JAMA Ophthalmology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":83,"is_retracted":false,"has_abstract":true,"ca_institutions":"Sunnybrook Health Science Centre; St. Michael's Hospital; McMaster University; University of Toronto","funders":"","keywords":"Chatbot; Medicine; Artificial intelligence; Medical physics; Ophthalmology; Optometry; Computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00127957,0.000153314,0.0005554719,0.0002918181,0.00003420976,0.00002292532,0.0001903689,0.0003393813,0.0007592984],"category_scores_gemma":[0.002573489,0.000137311,0.0002061321,0.0003961317,0.0003715968,0.0003278478,0.00004608942,0.0004466516,0.0001154549],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004006232,"about_ca_system_score_gemma":0.0005399527,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000557275,"about_ca_topic_score_gemma":0.00000492765,"domain_scores_codex":[0.9971536,0.0002698071,0.00167883,0.000425764,0.0002220523,0.0002499314],"domain_scores_gemma":[0.9972091,0.001474085,0.0003205885,0.0004327288,0.0004098141,0.0001537283],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001015669,0.0006476874,0.008127114,0.0007357518,0.00008289021,0.0003804532,0.002745165,0.00006470116,0.009363281,0.004013399,0.0001243194,0.9726996],"study_design_scores_gemma":[0.0001219747,0.01124375,0.05534973,0.00196942,0.0004097491,0.01074106,0.01172857,0.100979,0.6940862,0.1123083,0.0004341874,0.0006280696],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.989722,0.0009454652,0.002106741,0.001855518,0.002325575,0.0003755822,0.00001729857,0.00004561902,0.002606199],"genre_scores_gemma":[0.9967535,0.0001343542,0.002075288,0.0000863532,0.0007230126,0.0000248765,0.00005604856,0.00002530203,0.000121263],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9720715,"threshold_uncertainty_score":0.8313785,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2574401786806904,"score_gpt":0.5350129649838523,"score_spread":0.277572786303162,"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."}}