{"id":"W4409794823","doi":"10.2196/64963","title":"Comparing Diagnostic Accuracy of Clinical Professionals and Large Language Models: Systematic Review and Meta-Analysis","year":2025,"lang":"en","type":"review","venue":"JMIR Medical Informatics","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"CINAHL; Triage; MEDLINE; Meta-analysis; Medicine; Risk assessment; Computer science; Medical emergency; 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":["metaresearch","metaepi_broad"],"consensus_categories":[],"category_scores_codex":[0.04817605,0.002845417,0.01820138,0.01290457,0.0008876828,0.004408518,0.003805409,0.002650962,0.00371987],"category_scores_gemma":[0.1549565,0.001707626,0.03479706,0.01167412,0.00149484,0.004430369,0.001954782,0.002143621,0.0003115411],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003662389,"about_ca_system_score_gemma":0.006134902,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006955352,"about_ca_topic_score_gemma":0.01443121,"domain_scores_codex":[0.9575422,0.01995297,0.01336541,0.004145531,0.004359225,0.0006346183],"domain_scores_gemma":[0.7899173,0.1809821,0.01903895,0.003465056,0.005972869,0.0006238457],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"meta_analysis","study_design_scores_codex":[0.001564656,0.00004426836,0.01540894,0.5214487,0.4389598,0.000123898,0.0002198555,0.000568155,0.00008219967,0.0002175483,0.0007392748,0.02062283],"study_design_scores_gemma":[0.0006432682,0.0002518667,0.007297513,0.07793261,0.9107729,0.0001480021,0.000120271,0.0005680692,0.0001229557,0.0004933456,0.001595987,0.00005322957],"study_design_candidate":"meta_analysis","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.005153469,0.9918399,0.001181593,0.0003223653,0.0001276344,0.0004854923,0.0006538206,0.00002905424,0.0002066637],"genre_scores_gemma":[0.2634242,0.722669,0.006606258,0.00147298,0.0004909983,0.003378377,0.001682208,0.00005585119,0.0002201708],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9817986,"threshold_uncertainty_score":0.2547824,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.457867284644542,"score_gpt":0.5990253850888093,"score_spread":0.1411581004442674,"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."}}