{"id":"W4407274503","doi":"10.1093/jncics/pkaf021","title":"The use of large language models to enhance cancer clinical trial educational materials","year":2025,"lang":"en","type":"article","venue":"JNCI Cancer Spectrum","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Cancer Institute","keywords":"Readability; Automatic summarization; Clinical trial; Comprehension; Medicine; Plain language; Protocol (science); MEDLINE; Medical education; Alternative medicine; Computer science; Artificial intelligence; 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":[],"consensus_categories":[],"category_scores_codex":[0.01255376,0.001840519,0.0006008876,0.001671622,0.0005298445,0.001992365,0.001574579,0.001668339,0.00800002],"category_scores_gemma":[0.08165839,0.0005024672,0.001411341,0.0006393766,0.0006962718,0.00353383,0.003202989,0.002360562,0.003094369],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001672542,"about_ca_system_score_gemma":0.002171295,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001328533,"about_ca_topic_score_gemma":0.002830506,"domain_scores_codex":[0.988111,0.009433112,0.0005455329,0.001161448,0.0005931866,0.0001557294],"domain_scores_gemma":[0.9039033,0.08658899,0.002339977,0.003172076,0.003259729,0.0007358438],"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":[0.002198752,0.001218289,0.006723992,0.006030898,0.0002946356,0.0008287085,0.008886223,0.05725756,0.04030843,0.007519298,0.03756008,0.831173],"study_design_scores_gemma":[0.0009899406,0.002791669,0.005443277,0.001273472,0.0006158269,0.000886839,0.002392251,0.7641543,0.06518911,0.06756989,0.08833142,0.00036192],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08586966,0.00179736,0.8549106,0.005024714,0.0006390109,0.003223549,0.005581874,0.03698283,0.005970292],"genre_scores_gemma":[0.3606977,0.0004998088,0.6223085,0.001553932,0.0002163453,0.002745624,0.008087596,0.0008184264,0.00307213],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01255376,"threshold_uncertainty_score":0.06639147,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2787258091545623,"score_gpt":0.5640962906693894,"score_spread":0.2853704815148271,"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."}}