{"id":"W4281493375","doi":"10.3233/shti220403","title":"Pretrained Neural Networks Accurately Identify Cancer Recurrence in Medical Record","year":2022,"lang":"en","type":"article","venue":"Studies in health technology and informatics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"CancerCare Manitoba; University of Waterloo","funders":"","keywords":"Medical record; Computer science; Task (project management); Cancer; Artificial intelligence; Disease; Colorectal cancer; Domain (mathematical analysis); Breast cancer; Natural language processing; Medicine; Machine learning; Surgery; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008663791,0.000118306,0.0002705043,0.0002149724,0.0001992366,0.000004470357,0.0002676452,0.0002285331,0.00001511194],"category_scores_gemma":[0.0007530217,0.0001062239,0.00001845043,0.0005454222,0.0005551497,0.000005587196,0.0006722436,0.0006658271,2.904396e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000569625,"about_ca_system_score_gemma":0.0001238075,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004551881,"about_ca_topic_score_gemma":0.0004384741,"domain_scores_codex":[0.9986226,0.00008543964,0.0006173255,0.0001563254,0.0001439987,0.0003743025],"domain_scores_gemma":[0.9995284,0.00005704898,0.0001729808,0.0001629446,0.00002797254,0.00005062684],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001229943,0.00007191394,0.1197579,0.0004131033,0.00005367486,0.0000215336,0.002250846,0.0004209579,0.000006246117,0.0004988448,0.008073479,0.8683085],"study_design_scores_gemma":[0.0112746,0.0104083,0.08857285,0.001811127,0.00003845364,0.001052403,0.1563829,0.217954,0.0002345406,0.009172472,0.5005889,0.002509478],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9529755,0.03725553,0.0003620346,0.007961714,0.001014908,0.0002661792,0.00001308301,0.00004689534,0.0001042051],"genre_scores_gemma":[0.974286,0.02291718,0.0006631071,0.001832611,0.00003684283,0.0002141975,0.00001440202,0.00000541031,0.0000302178],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.865799,"threshold_uncertainty_score":0.4331683,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06847369989992134,"score_gpt":0.4168528629910206,"score_spread":0.3483791630910993,"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."}}