{"id":"W4395018673","doi":"10.2196/54283","title":"The Performance of ChatGPT-4V in Interpreting Images and Tables in the Japanese Medical Licensing Exam","year":2024,"lang":"en","type":"article","venue":"JMIR Medical Education","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Table (database); Computer science; Data mining; World Wide Web","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005115639,0.0007390572,0.0004836927,0.0007133299,0.0004374493,0.001327223,0.001013855,0.001069327,0.004973385],"category_scores_gemma":[0.02871407,0.000317848,0.0006958108,0.0003197829,0.0004465341,0.001686255,0.002105565,0.0008379974,0.002289381],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007179183,"about_ca_system_score_gemma":0.001358961,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007374358,"about_ca_topic_score_gemma":0.01227915,"domain_scores_codex":[0.9962386,0.0020821,0.0002333342,0.0006646529,0.0005400983,0.0002411528],"domain_scores_gemma":[0.9874578,0.008438766,0.0005040509,0.001019583,0.001879884,0.0006999496],"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":[0.004508104,0.00266994,0.1850693,0.001157379,0.0002461645,0.001159534,0.008569308,0.0119563,0.02991796,0.001067324,0.01237661,0.7413021],"study_design_scores_gemma":[0.0004772938,0.01531527,0.5278239,0.000693221,0.000600188,0.005018888,0.009949435,0.3327951,0.07457294,0.003843535,0.02829685,0.0006133163],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9815511,0.000212664,0.01038832,0.0002620029,0.00008890258,0.0002070628,0.0003213977,0.0009625012,0.006006157],"genre_scores_gemma":[0.9737355,0.0001422644,0.02117736,0.0002070204,0.00002856321,0.0001141896,0.0008027175,0.0001139365,0.003678543],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007374358,"threshold_uncertainty_score":0.02705443,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03461135957563634,"score_gpt":0.4288645519000784,"score_spread":0.3942531923244421,"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."}}