{"id":"W4376640706","doi":"10.1148/radiol.230987","title":"GPT-4 in Radiology: Improvements in Advanced Reasoning","year":2023,"lang":"en","type":"article","venue":"Radiology","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":130,"is_retracted":false,"has_abstract":true,"ca_institutions":"Women's College Hospital; Mount Sinai Hospital","funders":"","keywords":"Medicine; Medical physics; MEDLINE; Radiology; Nuclear medicine","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01182164,0.0008658164,0.001044207,0.001872568,0.0009250005,0.003657324,0.002900233,0.01002858,0.01414562],"category_scores_gemma":[0.06493671,0.000641219,0.002261455,0.001003247,0.002093457,0.005546944,0.001818309,0.01682015,0.005120033],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002328544,"about_ca_system_score_gemma":0.003313375,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004460228,"about_ca_topic_score_gemma":0.008560168,"domain_scores_codex":[0.9957665,0.00161807,0.0005437753,0.0003981094,0.001455877,0.0002176138],"domain_scores_gemma":[0.9444701,0.03618996,0.0009115181,0.001535334,0.01483074,0.002062323],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001516851,0.00003282517,0.0001697805,0.0008104791,0.000109865,0.000174197,0.00005444216,0.0003986188,0.0001924643,0.005101999,0.8853303,0.1074733],"study_design_scores_gemma":[0.0001533857,0.00007695622,0.00051182,0.001071385,0.0001782631,0.0006740771,0.00004516444,0.001403709,0.0004993762,0.01740661,0.9779349,0.00004450941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"commentary","genre_gemma":"empirical","genre_scores_codex":[0.0005938618,0.05451638,0.01079907,0.5764985,0.3493454,0.00005481986,0.0002978239,0.000653181,0.007241071],"genre_scores_gemma":[0.01323282,0.07972808,0.03445444,0.245564,0.6079139,0.0001469457,0.0009383805,0.0009678153,0.01705366],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01414562,"threshold_uncertainty_score":0.06251955,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09472364195722582,"score_gpt":0.4326889900903728,"score_spread":0.3379653481331469,"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."}}