{"id":"W3211439143","doi":"10.2196/27210","title":"A Question-and-Answer System to Extract Data From Free-Text Oncological Pathology Reports (CancerBERT Network): Development Study","year":2021,"lang":"en","type":"article","venue":"Journal of Medical Internet Research","topic":"Topic Modeling","field":"Computer Science","cited_by":42,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; Alberta Health Services","funders":"","keywords":"Computer science; Natural language processing; Artificial intelligence; Terminology; SNOMED CT; Named-entity recognition; Pathology; Deep learning; Information retrieval; Medicine; Linguistics","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":[0.01664329,0.0001326421,0.0004697326,0.0001429475,0.00008889118,0.0002725168,0.003266157,0.0002031651,0.0003610631],"category_scores_gemma":[0.004672625,0.00009676719,0.00004364171,0.0003515463,0.00007907549,0.0002666468,0.005941613,0.00149304,0.00002387257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003236026,"about_ca_system_score_gemma":0.001569209,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003351465,"about_ca_topic_score_gemma":0.001029644,"domain_scores_codex":[0.9927889,0.00154739,0.001195511,0.0007188545,0.003254178,0.0004951817],"domain_scores_gemma":[0.9958881,0.001152629,0.0002318255,0.001335634,0.0006733704,0.0007184953],"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":[0.0002409267,0.002054453,0.05499685,0.00008965832,0.0005738872,0.2243916,0.006659031,0.0004781658,0.0001361437,0.005260307,0.1687055,0.5364134],"study_design_scores_gemma":[0.005264267,0.004052001,0.1086039,0.006643165,0.0001052029,0.04190246,0.005122215,0.4751422,0.0005005989,0.005786457,0.3456931,0.00118435],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5307152,0.001775704,0.4549574,0.01005466,0.001781784,0.0002408138,0.000001361609,0.00003018467,0.0004429397],"genre_scores_gemma":[0.9240226,0.00007005173,0.07391617,0.0003699805,0.001239116,0.00001499397,0.000003246525,0.000009429768,0.0003544071],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5352291,"threshold_uncertainty_score":0.7405796,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1647109952524415,"score_gpt":0.4493167053224617,"score_spread":0.2846057100700201,"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."}}