{"id":"W4390415052","doi":"10.1016/j.jvsvi.2023.100049","title":"Evaluating the progression of artificial intelligence and large language models in medicine through comparative analysis of ChatGPT-3.5 and ChatGPT-4 in generating vascular surgery recommendations","year":2023,"lang":"en","type":"article","venue":"JVS-Vascular Insights","topic":"Cardiac, Anesthesia and Surgical Outcomes","field":"Medicine","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; University of Toronto","funders":"","keywords":"Computer science; Artificial intelligence; Medicine; Natural language processing; Linguistics; Cognitive science; Psychology; Philosophy","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":[0.02326148,0.001349545,0.0009770292,0.002196052,0.0005608944,0.002310671,0.001634346,0.002217691,0.002505233],"category_scores_gemma":[0.1143641,0.0005011255,0.001080383,0.0009182474,0.0008475265,0.002413878,0.003181845,0.002993949,0.001565854],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00200117,"about_ca_system_score_gemma":0.001646074,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005480438,"about_ca_topic_score_gemma":0.00631033,"domain_scores_codex":[0.9825643,0.01218089,0.001528341,0.001788484,0.00149857,0.000439447],"domain_scores_gemma":[0.7677724,0.2048771,0.005482128,0.007937889,0.009684278,0.004246196],"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":[0.0292363,0.009035072,0.1745728,0.006193231,0.002312209,0.0009979819,0.009749678,0.1512122,0.01927366,0.002518017,0.02376479,0.5711342],"study_design_scores_gemma":[0.002205967,0.01747223,0.1652626,0.0007030343,0.00106027,0.000763558,0.002606048,0.7676927,0.02417295,0.004582491,0.0129174,0.0005608198],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9712349,0.001459681,0.01433028,0.0008693841,0.0002840263,0.001023738,0.002818102,0.003335768,0.004644139],"genre_scores_gemma":[0.9430107,0.0003903715,0.04327689,0.0005351611,0.0001384838,0.001389861,0.00881632,0.0003252589,0.002116887],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02326148,"threshold_uncertainty_score":0.12302,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.151537108251437,"score_gpt":0.4222761822466601,"score_spread":0.2707390739952231,"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."}}