{"id":"W4224646058","doi":"10.18653/v1/2022.bionlp-1.32","title":"ICDBigBird: A Contextual Embedding Model for ICD Code Classification","year":2022,"lang":"en","type":"article","venue":"","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; University of Toronto","funders":"Mitacs","keywords":"Computer science; Word embedding; Task (project management); Embedding; Process (computing); Artificial intelligence; Health care; ICD-10; Machine learning; Graph; Natural language processing; Theoretical computer science","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002159082,0.000080297,0.00008998188,0.00002463274,0.0002010858,0.0000117184,0.0001742215,0.00007312755,0.00004367994],"category_scores_gemma":[0.0001141849,0.00007401654,0.0000696538,0.00004163861,0.00006264645,9.841509e-7,0.0001186002,0.00007066749,0.000002801017],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001752857,"about_ca_system_score_gemma":0.00006728217,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003713775,"about_ca_topic_score_gemma":0.00001449459,"domain_scores_codex":[0.9992897,0.00002880828,0.0001371266,0.0002607728,0.0001038729,0.0001797267],"domain_scores_gemma":[0.9996606,0.00002340822,0.00004710021,0.000186072,0.00003631136,0.00004651365],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005411727,0.000276911,0.0005244034,0.00004154952,0.0001272568,0.000001821918,0.0005228578,0.01127545,0.6577958,0.00956874,0.2358879,0.0834361],"study_design_scores_gemma":[0.000816172,0.0004105451,0.0001080126,0.000002418055,0.00001484408,0.00001080469,0.001403855,0.7343218,0.005019051,0.000391929,0.2573021,0.0001984865],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1776944,0.000544756,0.8157405,0.001918296,0.0002744497,0.0003348333,0.0002008815,0.0001102403,0.003181743],"genre_scores_gemma":[0.9741661,0.00001542883,0.01601001,0.0007280477,0.00008384523,0.0002246251,0.0002159769,0.00001482894,0.008541136],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7997304,"threshold_uncertainty_score":0.3018307,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06264871263909677,"score_gpt":0.3368751190488523,"score_spread":0.2742264064097555,"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."}}