{"id":"W4376640725","doi":"10.1148/radiol.230582","title":"Performance of ChatGPT on a Radiology Board-style Examination: Insights into Current Strengths and Limitations","year":2023,"lang":"en","type":"article","venue":"Radiology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":456,"is_retracted":false,"has_abstract":true,"ca_institutions":"Women's College Hospital","funders":"","keywords":"Medicine; Style (visual arts); Recall; Order (exchange); Radiology; Medical physics; Cognitive psychology; Psychology","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":[0.000216259,0.00008986682,0.000242018,0.0002909359,0.00009600701,0.000002478345,0.00004945915,0.0001150073,0.00002117177],"category_scores_gemma":[0.0007505863,0.00007849996,0.00002986438,0.0003029267,0.0001795361,0.00004617178,0.0000162878,0.0001637421,0.00008732641],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005478035,"about_ca_system_score_gemma":0.0001029731,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003751969,"about_ca_topic_score_gemma":0.00004036321,"domain_scores_codex":[0.9991024,0.0001119471,0.0003167179,0.0002121456,0.00007922268,0.0001775977],"domain_scores_gemma":[0.998551,0.0009842139,0.00008813408,0.0001784601,0.0001153667,0.0000828808],"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":[0.0001221587,0.0001193864,0.02589227,0.0002411565,0.00003069281,0.00000529253,0.0144348,0.00008405759,0.001965367,0.003632129,0.001919759,0.9515529],"study_design_scores_gemma":[0.000402022,0.00552289,0.8986038,0.0003379086,0.0000923351,0.0001485733,0.004619653,0.03297151,0.009414388,0.002348901,0.04521889,0.0003191401],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9952875,0.001437887,0.00002999715,0.001449665,0.0007598909,0.0002264983,0.000001100034,0.00004820845,0.0007592685],"genre_scores_gemma":[0.9902776,0.00891857,0.0001012337,0.0001454623,0.0003090754,0.00005316562,0.00006260861,0.000008826062,0.0001234531],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9512338,"threshold_uncertainty_score":0.3201135,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1150913827002931,"score_gpt":0.3991779660783883,"score_spread":0.2840865833780953,"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."}}