{"id":"W4283687125","doi":"10.2196/37557","title":"Automatic International Classification of Diseases Coding System: Deep Contextualized Language Model With Rule-Based Approaches","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Medical Coding and Health Information","field":"Health Professions","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Far Eastern Memorial Hospital; Ministry of Science and Technology, Taiwan","keywords":"Computer science; Medical diagnosis; Artificial intelligence; Medical classification; Diagnosis code; Unified Medical Language System; Preprocessor; Word2vec; Coding (social sciences); Language model; Natural language processing; Word embedding; Data mining; Machine learning; Medicine; Embedding","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001591699,0.0001590124,0.0003959564,0.0002427414,0.0006625561,0.00001284389,0.0004546221,0.0001590155,0.001195466],"category_scores_gemma":[0.0004352023,0.0001208985,0.00006412597,0.0002711514,0.0001363688,0.000248063,0.0001494834,0.0008356975,0.00004676202],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000443908,"about_ca_system_score_gemma":0.001214599,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001986646,"about_ca_topic_score_gemma":0.00000551363,"domain_scores_codex":[0.9956414,0.0003085719,0.001640762,0.0001014672,0.001938886,0.0003689367],"domain_scores_gemma":[0.9974857,0.0005916334,0.00107219,0.0003075502,0.0001546668,0.0003882377],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002342184,0.001930851,0.02588549,0.07753206,0.0005019041,0.00004234977,0.4110457,0.01777612,0.00004021206,0.2345855,0.04250843,0.1858092],"study_design_scores_gemma":[0.002266973,0.00007891694,0.00081212,0.0006197599,0.00002935183,0.000005473533,0.07807226,0.9170814,0.000002414024,0.0000272206,0.0008848507,0.0001192707],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6354356,0.00009458489,0.3133383,0.003995696,0.0009779601,0.003345748,0.0003821578,0.0008953036,0.04153454],"genre_scores_gemma":[0.9916452,0.000004502316,0.004200173,0.0020297,0.00009065691,0.001195292,0.0006674631,0.00001718265,0.0001498031],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8993053,"threshold_uncertainty_score":0.9997176,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1839677208632765,"score_gpt":0.4185987175406165,"score_spread":0.23463099667734,"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."}}