{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002483763,0.0001768227,0.001337008,0.0006350731,0.00009058056,0.00001178232,0.00006482092,0.00009934092,0.00002616241],"category_scores_gemma":[0.0003942131,0.0001109694,0.0003084904,0.00218582,0.0001745223,0.0001345772,0.00007103901,0.0001907022,8.489212e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002256515,"about_ca_system_score_gemma":0.00004638729,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003577592,"about_ca_topic_score_gemma":0.0005308753,"domain_scores_codex":[0.9975342,0.0005579968,0.0008063028,0.0003370039,0.0005352897,0.0002291634],"domain_scores_gemma":[0.9979019,0.001392144,0.000203804,0.0003354761,0.000108567,0.00005809329],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006381329,0.001343349,0.1220616,0.001335077,0.007055353,0.0004726166,0.5042256,0.05296143,0.01713775,0.05877233,0.00007669965,0.2339201],"study_design_scores_gemma":[0.0008829213,0.0002245807,0.06702372,0.0009230776,0.001424889,0.000009708659,0.01891214,0.9073021,0.001641086,0.0007886579,0.0006007583,0.0002663819],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9911736,0.006015092,0.001381919,0.0006397482,0.00004698765,0.0005177143,0.000004303442,0.00001999066,0.0002006383],"genre_scores_gemma":[0.9969172,0.001815741,0.0009350743,0.0001004276,0.00005182963,0.00004778235,0.0001080351,0.00001095922,0.00001289828],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8543407,"threshold_uncertainty_score":0.4525202,"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."}}