{"id":"W7117468468","doi":"10.3390/bioengineering13010030","title":"Accurate Clinical Entity Recognition and Code Mapping of Anatomopathological Reports Using BioClinicalBERT Enhanced by Retrieval-Augmented Generation: A Hybrid Deep Learning Approach","year":2025,"lang":"en","type":"article","venue":"Bioengineering","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Deep learning; Pipeline (software); Coding (social sciences); Code (set theory); Pattern recognition (psychology)","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":[0.001907216,0.001402376,0.000759215,0.004885327,0.000534338,0.001477542,0.002026862,0.001324792,0.003101839],"category_scores_gemma":[0.003883359,0.0003806681,0.001433766,0.002067807,0.0005177204,0.001609543,0.001933495,0.00113722,0.002706965],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001494158,"about_ca_system_score_gemma":0.002044643,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01383473,"about_ca_topic_score_gemma":0.01910557,"domain_scores_codex":[0.9987823,0.0002040522,0.0001233176,0.0004778664,0.0002837085,0.0001287049],"domain_scores_gemma":[0.9984712,0.0003853978,0.0001881927,0.000387106,0.0005066313,0.00006149805],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0003273175,0.0003037372,0.01105804,0.000501348,0.0002299166,0.0005181759,0.0003229185,0.02367988,0.02563798,0.001824994,0.03977555,0.8958202],"study_design_scores_gemma":[0.0001042183,0.0002615342,0.01621468,0.0002479957,0.0003775033,0.00123864,0.0003912396,0.879095,0.05639872,0.008855182,0.03669104,0.0001243965],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1593019,0.003925799,0.7334996,0.001490338,0.0004987496,0.0008666171,0.01856821,0.07474169,0.007107104],"genre_scores_gemma":[0.3732145,0.001012974,0.5599388,0.000725329,0.0001971141,0.0005002179,0.05537606,0.0009443578,0.008090732],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01383473,"threshold_uncertainty_score":0.02750844,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06099622682867463,"score_gpt":0.3221194033208158,"score_spread":0.2611231764921412,"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."}}