{"id":"W2912907847","doi":"10.1016/j.jbi.2019.103114","title":"Automatic ICD code assignment of Chinese clinical notes based on multilayer attention BiRNN","year":2019,"lang":"en","type":"article","venue":"Journal of Biomedical Informatics","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":66,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Saskatchewan","funders":"Higher Education Discipline Innovation Project; National Natural Science Foundation of China","keywords":"Computer science; Code (set theory); Feature engineering; Word embedding; Feature (linguistics); Artificial intelligence; Natural language processing; Block (permutation group theory); Representation (politics); Artificial neural network; Semantics (computer science); Word (group theory); Recurrent neural network; Hamming code; Embedding; Deep learning; Block code; Linguistics; Programming language; Algorithm; Decoding methods","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001014123,0.0005741303,0.0006166988,0.005476867,0.0009626935,0.0008770082,0.000793725,0.0006075577,0.002029266],"category_scores_gemma":[0.003711454,0.0001880617,0.0006142591,0.002491791,0.0002550845,0.0008955427,0.001103438,0.0006961726,0.001013055],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009972298,"about_ca_system_score_gemma":0.002159161,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02253406,"about_ca_topic_score_gemma":0.02781462,"domain_scores_codex":[0.9987773,0.0001587451,0.0002278451,0.0003734725,0.0002717872,0.0001907575],"domain_scores_gemma":[0.9976894,0.0007535522,0.0002156507,0.0001871766,0.001021933,0.0001324431],"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":[0.0009337997,0.0005439867,0.1156458,0.0007332645,0.0001607742,0.001219408,0.0006723348,0.006452632,0.0319125,0.002164459,0.03313035,0.8064307],"study_design_scores_gemma":[0.0001789391,0.0003749518,0.1641747,0.00033888,0.0008496188,0.002201351,0.001687102,0.760712,0.03653361,0.01025129,0.02251406,0.0001834514],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7094874,0.004876406,0.2473387,0.002256937,0.001159214,0.0008139049,0.01805188,0.00735548,0.008660165],"genre_scores_gemma":[0.8466259,0.00112701,0.1209034,0.0003374996,0.0002855106,0.0003306027,0.02628488,0.0001260712,0.003979124],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02253406,"threshold_uncertainty_score":0.04480582,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02064373706993933,"score_gpt":0.3388898037443068,"score_spread":0.3182460666743675,"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."}}