{"id":"W4406865074","doi":"10.2196/65984","title":"Large Language Model Applications for Health Information Extraction in Oncology: Scoping Review","year":2025,"lang":"en","type":"review","venue":"JMIR Cancer","topic":"Topic Modeling","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"BC Cancer Agency; University of Waterloo; University of Toronto","funders":"","keywords":"Preprint; Computer science; Data science; Medicine; World Wide Web","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"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":[0.03979845,0.002027066,0.004506119,0.02335501,0.001280664,0.005248263,0.003821226,0.003326264,0.00672504],"category_scores_gemma":[0.2048238,0.001515594,0.009605555,0.02225048,0.002204617,0.006641968,0.004031536,0.002974833,0.001555648],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005986745,"about_ca_system_score_gemma":0.02464084,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01081438,"about_ca_topic_score_gemma":0.01734942,"domain_scores_codex":[0.9811767,0.008528094,0.006277231,0.001131188,0.002647329,0.0002395071],"domain_scores_gemma":[0.7355708,0.2345575,0.01216177,0.004091359,0.01307006,0.0005484893],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"systematic_review","study_design_scores_codex":[0.0001428992,0.00006304,0.0009969189,0.6133459,0.002289987,0.0001588051,0.0009067745,0.00127468,0.0002667901,0.003393642,0.007519396,0.3696411],"study_design_scores_gemma":[0.00005425524,0.0001410955,0.001616831,0.9009079,0.007238428,0.0002827754,0.0006628515,0.000872555,0.0005297211,0.003871349,0.08375169,0.00007043777],"study_design_candidate":"systematic_review","study_design_consensus":"systematic_review","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.0005023931,0.9917073,0.003735296,0.001363797,0.000203219,0.0007425489,0.0006339188,0.00007265898,0.001038854],"genre_scores_gemma":[0.006767063,0.9796935,0.009393018,0.0008123659,0.0001791956,0.002095173,0.0008420044,0.00004349154,0.0001742328],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.03979845,"threshold_uncertainty_score":0.2104769,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.074606718448783,"score_gpt":0.5065021944205048,"score_spread":0.4318954759717218,"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."}}