{"id":"W4407119832","doi":"10.1097/js9.0000000000002215","title":"AI-driven evidence synthesis: data extraction of randomized controlled trials with large language models","year":2025,"lang":"en","type":"article","venue":"International Journal of Surgery","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University; Impact","funders":"Fundamental Research Funds for the Central Universities; China Academy of Chinese Medical Sciences; National Natural Science Foundation of China","keywords":"Data extraction; Randomized controlled trial; Medicine; Meta-analysis; Systematic review; MEDLINE; Artificial intelligence; Data mining; Computer science; Natural language processing; Pathology","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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.3705406,0.005951557,0.009313101,0.01524892,0.002019698,0.008202433,0.004545287,0.003859933,0.01837675],"category_scores_gemma":[0.7427641,0.004379395,0.01408746,0.01226628,0.002412366,0.007900292,0.009022743,0.006181356,0.004733803],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005632639,"about_ca_system_score_gemma":0.02690004,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002369774,"about_ca_topic_score_gemma":0.004131378,"domain_scores_codex":[0.4797623,0.4374478,0.05925237,0.01101913,0.01188814,0.0006302294],"domain_scores_gemma":[0.1185454,0.8222014,0.02218316,0.02560782,0.01071946,0.0007427534],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.006880009,0.0002640563,0.00345517,0.2471503,0.01537322,0.0005691529,0.005970456,0.02400801,0.003244216,0.04769547,0.034729,0.610661],"study_design_scores_gemma":[0.01549476,0.00198699,0.004087127,0.0792849,0.0198115,0.000732631,0.001707886,0.1711722,0.01517516,0.5497446,0.1394527,0.001349569],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003180223,0.009483973,0.9374546,0.005327792,0.0007980207,0.02741718,0.008318172,0.006069554,0.001950439],"genre_scores_gemma":[0.02367151,0.001809397,0.9275915,0.001232092,0.0002074276,0.04307181,0.001777537,0.0003136674,0.0003249938],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6294594,"threshold_uncertainty_score":0.7762355,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.310832927540387,"score_gpt":0.5076900762546582,"score_spread":0.1968571487142712,"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."}}