{"id":"W4409657910","doi":"10.1097/rct.0000000000001709","title":"Beyond Human Limits: The Promise and Pitfalls of Large Language Models in Radiology Research","year":2025,"lang":"en","type":"review","venue":"Journal of Computer Assisted Tomography","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto General Hospital","funders":"","keywords":"Medicine; Productivity; Engineering ethics; Data science; Computer science","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":[0.002713177,0.0001890364,0.001395608,0.001895624,0.00009665868,0.00002767046,0.0003045682,0.0003050084,0.000009591373],"category_scores_gemma":[0.0001032336,0.0001159901,0.0004536113,0.001296768,0.0001801097,0.0000757532,0.00006771774,0.001355589,6.357795e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007372599,"about_ca_system_score_gemma":0.0006992328,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000594806,"about_ca_topic_score_gemma":0.00004247329,"domain_scores_codex":[0.9968175,0.000892466,0.001427827,0.0002032965,0.0003593886,0.0002994774],"domain_scores_gemma":[0.9971882,0.001162926,0.0006331771,0.000335932,0.0005621978,0.0001176289],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00004894962,0.0003740768,0.0003709906,0.01124828,0.0002716549,0.00007306986,0.00254001,0.000002403164,0.000004151408,0.0006349452,0.002131364,0.9823001],"study_design_scores_gemma":[0.001897637,0.00855849,0.007689558,0.2804067,0.004593851,0.004832231,0.005577657,0.0007694812,0.0001107946,0.02000118,0.6642579,0.001304531],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.004466132,0.9932873,0.0002758978,0.0004434817,0.0003948049,0.0007504601,0.000009137187,0.000005365524,0.0003673779],"genre_scores_gemma":[0.009287695,0.9889986,0.0008581764,0.000126585,0.0006181867,0.00002229054,0.00001700562,0.00001726767,0.00005419386],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9809956,"threshold_uncertainty_score":0.5889434,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2145296303470314,"score_gpt":0.504344560250096,"score_spread":0.2898149299030646,"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."}}