{"id":"W3182655775","doi":"10.2196/27970","title":"A Natural Language Processing–Assisted Extraction System for Gleason Scores: Development and Usability Study","year":2021,"lang":"en","type":"article","venue":"JMIR Cancer","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Cancer Institute; Abramson Cancer Center; University of Pennsylvania Health System; University of Pennsylvania","keywords":"Artificial intelligence; Medicine; Prostate cancer; Cohort; Natural language processing; Categorization; Computer science; Cancer; Pathology; Internal medicine","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001046861,0.00008605369,0.0001150744,0.00001109648,0.00008907516,0.00003060415,0.00004784696,0.00008362914,0.000005009492],"category_scores_gemma":[0.00006777963,0.0000695336,0.0000260456,0.00005632409,0.00004313907,0.000002595424,0.00004061368,0.00005451082,4.89427e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004486416,"about_ca_system_score_gemma":0.0001756325,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002711732,"about_ca_topic_score_gemma":0.0004738132,"domain_scores_codex":[0.9993395,0.00003130471,0.0001246226,0.0002892055,0.00008523768,0.0001301397],"domain_scores_gemma":[0.9997045,0.00001172558,0.0000503096,0.000113118,0.00008078914,0.00003959811],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0002385669,0.0002588462,0.02169429,0.0005769039,0.00008651006,0.00001446539,0.001497889,8.427663e-7,0.1607579,0.000001937083,0.0009259502,0.8139459],"study_design_scores_gemma":[0.003267236,0.0005122417,0.5901535,0.0003814243,0.00008423446,0.0000752233,0.02768085,0.0003501585,0.255141,0.000002503954,0.1217655,0.000586122],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9885098,0.01055621,0.0003163834,0.00007869751,0.0001764117,0.0002794075,0.000007115866,0.00002707218,0.00004888879],"genre_scores_gemma":[0.996937,0.00001075653,0.001799869,0.00004406609,0.0001336456,0.0004112553,0.00003278964,0.000007761554,0.0006228519],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8133598,"threshold_uncertainty_score":0.2835498,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02289925283352091,"score_gpt":0.3524482616514882,"score_spread":0.3295490088179673,"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."}}