{"id":"W4407135853","doi":"10.1148/radiol.242134","title":"Leveraging Large Language Models to Generate Clinical Histories for Oncologic Imaging Requisitions","year":2025,"lang":"en","type":"article","venue":"Radiology","topic":"Radiomics and Machine Learning in Medical Imaging","field":"Medicine","cited_by":35,"is_retracted":false,"has_abstract":true,"ca_institutions":"Princess Margaret Cancer Centre","funders":"","keywords":"Requisition; Medicine; Interpretation (philosophy); Medical physics; Radiology; Cancer imaging; Cancer; Linguistics; Archaeology; Internal medicine","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":[0.001175138,0.0001358192,0.0004814219,0.0001868456,0.0001934036,0.0000159533,0.0001329997,0.0001022304,0.00003736507],"category_scores_gemma":[0.0009323516,0.0001195937,0.0001552107,0.000164579,0.000121707,0.00004635988,0.00007394699,0.0003870425,0.000009459016],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002086983,"about_ca_system_score_gemma":0.0002126368,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002750888,"about_ca_topic_score_gemma":0.000004453585,"domain_scores_codex":[0.9985319,0.0001342129,0.0004070463,0.0003942489,0.00007210339,0.0004605045],"domain_scores_gemma":[0.9989867,0.0004062646,0.00006414607,0.0002798693,0.00008789209,0.0001751301],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00105654,0.0007859159,0.06755119,0.0003991592,0.0006864326,0.0007601483,0.007279617,0.005076529,0.02254234,0.1690038,0.5053425,0.2195158],"study_design_scores_gemma":[0.008320845,0.0005130657,0.01546556,0.0002377799,0.0005085574,0.0004924394,0.001767731,0.4242241,0.0002344063,0.01623031,0.5314333,0.0005718927],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3562319,0.001991878,0.5971098,0.03494502,0.001973195,0.0005790138,0.00001366216,0.0001932912,0.006962292],"genre_scores_gemma":[0.930578,0.00006484659,0.03716192,0.02702262,0.0007597733,0.00009304145,0.00008043832,0.0000229918,0.004216362],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5743461,"threshold_uncertainty_score":0.487689,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03855689954147529,"score_gpt":0.3957586267377718,"score_spread":0.3572017271962966,"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."}}