{"id":"W4414759776","doi":"10.2196/72005","title":"Cancer Diagnosis Categorization in Electronic Health Records Using Large Language Models and BioBERT: Model Performance Evaluation Study","year":2025,"lang":"en","type":"article","venue":"JMIR Cancer","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Documentation; Categorization; Health records; Cancer; Electronic health record; MEDLINE; Data collection; Health informatics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009356322,0.0001530007,0.0002164301,0.0002731162,0.0001973307,0.00006758268,0.0002683531,0.00006214569,0.00001835145],"category_scores_gemma":[0.00001946594,0.000154357,0.00001877625,0.0008842035,0.00001068817,0.0005328018,0.0001417954,0.0003025577,5.56172e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001411564,"about_ca_system_score_gemma":0.001551786,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.01508553,"about_ca_topic_score_gemma":0.02189797,"domain_scores_codex":[0.9981359,0.0002606533,0.0003060904,0.0005053606,0.0003241912,0.0004678437],"domain_scores_gemma":[0.9993021,0.00003177306,0.00014352,0.0003427103,0.0001245286,0.00005538112],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001211498,0.000106683,0.4234401,0.0001277577,0.00001104601,4.498234e-7,0.008823181,0.4133695,0.000009020288,0.001304211,0.0001277546,0.1526682],"study_design_scores_gemma":[0.000638572,0.00007253756,0.05680443,0.000138209,0.000007459238,3.162859e-7,0.0001369448,0.9415272,0.00001782693,0.0004591425,0.00006926699,0.000128127],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9358153,0.006808274,0.05394775,0.001642131,0.0002511789,0.001390318,0.000008524898,0.00007504212,0.00006150454],"genre_scores_gemma":[0.9946528,0.001547085,0.000660546,0.000714681,0.00004298521,0.002270573,0.000005713131,0.0000135466,0.0000920911],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5281577,"threshold_uncertainty_score":0.9959499,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0396032572425928,"score_gpt":0.4156220820246641,"score_spread":0.3760188247820713,"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."}}