{"id":"W4416217660","doi":"10.2196/76912","title":"Named Entity Recognition for Chinese Cancer Electronic Health Records—Development and Evaluation of a Domain-Specific BERT Model: Quantitative Study","year":2025,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Named-entity recognition; Breast cancer; Cancer; Clinical decision support system; Electronic health record; Cancer treatment","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01054046,0.0012724,0.0007768595,0.002082923,0.0005755787,0.001278164,0.001609668,0.0009856852,0.001521137],"category_scores_gemma":[0.01847221,0.0003259972,0.0008196988,0.001868829,0.0005802037,0.004215028,0.001180088,0.001127714,0.0005469744],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00291379,"about_ca_system_score_gemma":0.001652037,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03612575,"about_ca_topic_score_gemma":0.02517604,"domain_scores_codex":[0.996527,0.001592153,0.0002526931,0.000764204,0.0006778702,0.0001860781],"domain_scores_gemma":[0.9877695,0.008292748,0.0004986087,0.001248485,0.00193448,0.0002561871],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001882545,0.0019815,0.08385187,0.001833861,0.0008191273,0.0005919611,0.0008668383,0.2698372,0.008951535,0.004539301,0.0235798,0.6012646],"study_design_scores_gemma":[0.00004122885,0.0004229165,0.02082951,0.00006381724,0.0002113362,0.0001613171,0.0003526161,0.9685102,0.005659652,0.0008459822,0.002853514,0.00004788395],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9158265,0.004481791,0.0645197,0.0009465934,0.000246357,0.0006598605,0.005358717,0.002359296,0.005601164],"genre_scores_gemma":[0.9506994,0.001272271,0.03353858,0.0001062318,0.00006450063,0.0002325215,0.01227076,0.00006355112,0.001752235],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03612575,"threshold_uncertainty_score":0.07183093,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06409140617425216,"score_gpt":0.3993231893029531,"score_spread":0.3352317831287009,"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."}}