{"id":"W2612992329","doi":"10.2196/medinform.7235","title":"Effective Information Extraction Framework for Heterogeneous Clinical Reports Using Online Machine Learning and Controlled Vocabularies","year":2017,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Information extraction; Controlled vocabulary; Machine learning; Artificial intelligence; Unified Medical Language System; Information retrieval; Data science; Data mining","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.001041911,0.000167133,0.0003886873,0.00004172327,0.0004317912,0.0001435454,0.0001539025,0.0006757783,0.00001075201],"category_scores_gemma":[0.009494543,0.000128561,0.0001364011,0.00002154757,0.0004434087,0.00003543632,0.000185714,0.0004206282,0.000001643568],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001255738,"about_ca_system_score_gemma":0.00009638676,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001043386,"about_ca_topic_score_gemma":0.000009907022,"domain_scores_codex":[0.9984186,0.00006653806,0.0008594251,0.0001268919,0.0002918366,0.0002367489],"domain_scores_gemma":[0.9982959,0.0002616279,0.0008163319,0.0002925758,0.0001258968,0.0002076466],"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":[0.002299069,0.0003028394,0.03495545,0.0005148784,0.0005751537,0.0000370202,0.001138021,0.00005409399,0.0003394752,0.0002233895,0.0006434633,0.9589171],"study_design_scores_gemma":[0.0232075,0.006151109,0.02003811,0.001006326,0.0004645591,0.002208256,0.00279514,0.4201597,0.003073788,0.003435155,0.5158229,0.001637391],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8189875,0.0003261505,0.1789679,0.0002774399,0.0005702497,0.0007302469,0.00001792158,0.00004182626,0.0000806941],"genre_scores_gemma":[0.9507379,0.000362888,0.04732457,0.0006545123,0.0005163719,0.00009491706,0.0002577945,0.0000132876,0.00003775987],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9572797,"threshold_uncertainty_score":0.9988489,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02581817604947437,"score_gpt":0.3984720093010376,"score_spread":0.3726538332515632,"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."}}