{"id":"W4387305434","doi":"10.2196/48023","title":"Accuracy of ChatGPT on Medical Questions in the National Medical Licensing Examination in Japan: Evaluation Study","year":2023,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":106,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Chiba University","keywords":"Medical diagnosis; Christian ministry; Welfare; Reliability (semiconductor); Medical education; Psychology; Medicine; Political science; Law; Pathology","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":[0.01421652,0.000833515,0.001128337,0.001973245,0.0006610961,0.001165523,0.001100138,0.001521663,0.001747297],"category_scores_gemma":[0.08229294,0.0003777883,0.001366176,0.001168894,0.0007698588,0.001970193,0.001978308,0.0009566611,0.00117221],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00122701,"about_ca_system_score_gemma":0.001178435,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00786192,"about_ca_topic_score_gemma":0.007287984,"domain_scores_codex":[0.9878979,0.006163512,0.001707974,0.001513818,0.002266769,0.0004500219],"domain_scores_gemma":[0.9081045,0.05185332,0.009407094,0.004643532,0.02096029,0.005031237],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.003783496,0.001452028,0.8852143,0.001015529,0.0003177613,0.0004975027,0.008368097,0.001077885,0.001803899,0.00009822182,0.003132385,0.09323883],"study_design_scores_gemma":[0.0002724602,0.003184699,0.9646569,0.0002719993,0.0006132213,0.0009303571,0.003533828,0.0203843,0.002891915,0.0001594885,0.0029874,0.0001135105],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9957556,0.000470055,0.001176031,0.0001413014,0.00004906524,0.0003379802,0.0005661806,0.0001302911,0.001373595],"genre_scores_gemma":[0.9941883,0.0003687145,0.002800218,0.000120276,0.00006298362,0.000359248,0.001193979,0.00003182159,0.0008744405],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01421652,"threshold_uncertainty_score":0.07518506,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5099773349532368,"score_gpt":0.6438245307259427,"score_spread":0.1338471957727059,"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."}}