{"id":"W4408158553","doi":"10.1200/cci-24-00143","title":"Using a Longformer Large Language Model for Segmenting Unstructured Cancer Pathology Reports","year":2025,"lang":"en","type":"article","venue":"JCO Clinical Cancer Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; University of British Columbia","funders":"","keywords":"Pathology; Market segmentation; Computer science; Natural language processing; Medicine; Artificial intelligence; Business","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001221817,0.0001932617,0.0004319746,0.0001125616,0.0001896091,0.0001236805,0.0005240152,0.000230779,0.00001387632],"category_scores_gemma":[0.0002927259,0.0001715003,0.000195768,0.0002730622,0.00004888702,0.0005719828,0.0004380438,0.0003245661,0.00000125816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001466526,"about_ca_system_score_gemma":0.0006873324,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006321726,"about_ca_topic_score_gemma":0.0001595144,"domain_scores_codex":[0.9972652,0.00003400373,0.00162757,0.0003019182,0.0002424013,0.0005288825],"domain_scores_gemma":[0.9981529,0.0001742563,0.0006233998,0.0007516147,0.0002002458,0.00009759517],"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":[0.00008990667,0.0002031202,0.05500611,0.001636576,0.0003611072,0.00006864186,0.02218926,0.4382537,0.0002849532,0.03578809,0.005927537,0.440191],"study_design_scores_gemma":[0.0006792322,0.00001306444,0.0002146882,0.0001129648,0.00004779444,0.00001040042,0.0001821188,0.9937961,0.0001356623,0.002796365,0.001820601,0.0001910766],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1144497,0.000329728,0.8819286,0.0002872968,0.001919966,0.000407264,0.00002960968,0.00011495,0.0005327954],"genre_scores_gemma":[0.4075764,0.0001048621,0.5836363,0.0074699,0.0002900509,0.0001500659,0.0000104374,0.00001772768,0.0007443346],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.5555423,"threshold_uncertainty_score":0.6993577,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0939272117815288,"score_gpt":0.4464086519327034,"score_spread":0.3524814401511746,"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."}}