{"id":"W4414161315","doi":"10.2196/68707","title":"Performance of Natural Language Processing for Information Extraction From Electronic Health Records Within Cancer: Systematic Review","year":2025,"lang":"en","type":"review","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Aalborg Universitetshospital; Aalborg Universitet","keywords":"Information extraction; Health records; Electronic health record; Natural language; Unstructured data; Biomedical text mining; Text mining; Text processing; Information processing","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":[0.04375592,0.002177309,0.008867646,0.01981188,0.0009822316,0.003926158,0.003093972,0.002086956,0.00245236],"category_scores_gemma":[0.2172116,0.001392075,0.01754912,0.01419819,0.001660868,0.005922467,0.002310764,0.001598203,0.0003745931],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005576495,"about_ca_system_score_gemma":0.01750159,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01093482,"about_ca_topic_score_gemma":0.02380796,"domain_scores_codex":[0.9627238,0.01495158,0.01329095,0.002766471,0.00582998,0.0004371833],"domain_scores_gemma":[0.716673,0.2442717,0.02076201,0.00269248,0.01496072,0.0006401238],"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":[0.0003272248,0.00003005719,0.004154813,0.9093565,0.02343861,0.0001028256,0.0003006678,0.0004192294,0.000150703,0.0001863218,0.0009967444,0.06053639],"study_design_scores_gemma":[0.000292906,0.0004958981,0.01034681,0.7896711,0.1846046,0.0005147345,0.0005089827,0.0009248023,0.000696541,0.0006242734,0.01121296,0.0001065032],"study_design_candidate":"systematic_review","study_design_consensus":"systematic_review","genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.003480756,0.9928232,0.001088838,0.000515416,0.0001352795,0.0008353451,0.0007529342,0.00003573321,0.0003324553],"genre_scores_gemma":[0.06719124,0.922034,0.006436594,0.0008975794,0.0001848142,0.001861191,0.001226961,0.00003285431,0.000134838],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.04375592,"threshold_uncertainty_score":0.2314062,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01986415516832182,"score_gpt":0.3726917215292588,"score_spread":0.352827566360937,"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."}}