{"id":"W4404518252","doi":"10.1200/cci.24.00110","title":"Classifying Tumor Reportability Status From Unstructured Electronic Pathology Reports Using Language Models in a Population-Based Cancer Registry Setting","year":2024,"lang":"en","type":"article","venue":"JCO Clinical Cancer Informatics","topic":"AI in cancer detection","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Provincial Health Services Authority; University of British Columbia","funders":"","keywords":"Medical diagnosis; Artificial intelligence; Population; Computer science; Cancer registry; Pipeline (software); Natural language processing; Machine learning; Cancer; Set (abstract data type); Medicine; Medical physics; Pathology; Internal medicine","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":[0.003411258,0.0007487087,0.0004010801,0.00199291,0.0002636975,0.001031326,0.0009472051,0.000611449,0.0008035505],"category_scores_gemma":[0.0115233,0.0002802767,0.000664693,0.000974397,0.0002774558,0.0008476789,0.0009376416,0.001268609,0.0007101673],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001084295,"about_ca_system_score_gemma":0.001681964,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02486327,"about_ca_topic_score_gemma":0.02613151,"domain_scores_codex":[0.9986994,0.0005461854,0.0001167516,0.0003481296,0.0001911655,0.00009820239],"domain_scores_gemma":[0.9927351,0.004806093,0.0007687699,0.0005763043,0.0008425968,0.0002711504],"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":[0.001058826,0.001339131,0.5182506,0.0003405458,0.0004706655,0.001178452,0.0007357158,0.1268576,0.005764187,0.0007149538,0.01872537,0.3245639],"study_design_scores_gemma":[0.00008956149,0.000319707,0.07269275,0.00008193815,0.0001775912,0.0004883448,0.0006062828,0.9123724,0.007154802,0.002376268,0.00356813,0.0000722121],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9412931,0.0004977601,0.04461336,0.001843412,0.00009809325,0.0002027811,0.006652762,0.003171637,0.001627182],"genre_scores_gemma":[0.9550269,0.0001979271,0.03298406,0.000308397,0.00007390474,0.0001303833,0.01027844,0.00005616283,0.0009438346],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02486327,"threshold_uncertainty_score":0.04943711,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04502848308434197,"score_gpt":0.3846905372981353,"score_spread":0.3396620542137934,"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."}}