{"id":"W3156931326","doi":"10.2196/22461","title":"Neural Machine Translation–Based Automated Current Procedural Terminology Classification System Using Procedure Text: Development and Validation Study","year":2021,"lang":"en","type":"article","venue":"JMIR Formative Research","topic":"Medical Coding and Health Information","field":"Health Professions","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of General Medical Sciences; Medical School, University of Michigan; National Institutes of Health","keywords":"Computer science; Artificial intelligence; Machine learning; Medical diagnosis; Support vector machine; Preprocessor; Machine translation; Data mining; Data pre-processing; Terminology; Medical classification; Medicine","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002190435,0.0008778492,0.0005092031,0.0009899687,0.0003999919,0.0005512856,0.001207598,0.0007337267,0.001716521],"category_scores_gemma":[0.005789472,0.0002406085,0.0007636689,0.0008783244,0.0003943996,0.001036882,0.0006821075,0.0007661825,0.001234229],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001193255,"about_ca_system_score_gemma":0.001416555,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01017339,"about_ca_topic_score_gemma":0.008052334,"domain_scores_codex":[0.9988002,0.0003447561,0.0001490979,0.000283021,0.0003340573,0.00008888002],"domain_scores_gemma":[0.9960455,0.001443221,0.0002928528,0.0004575342,0.001648978,0.0001118974],"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":[0.001665732,0.005847483,0.07581132,0.0009401552,0.0006464227,0.00125157,0.0006585289,0.1524386,0.03433679,0.0009598165,0.01456398,0.7108797],"study_design_scores_gemma":[0.0001785095,0.001941116,0.040278,0.00006244791,0.0001638295,0.0004728139,0.0002195614,0.92968,0.02301019,0.000363182,0.003578113,0.00005229509],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.970089,0.0005365237,0.02377323,0.0002187752,0.0001368248,0.0004851601,0.001350738,0.001453222,0.001956595],"genre_scores_gemma":[0.9333854,0.0004216365,0.05260127,0.0001695567,0.00006028808,0.0004961303,0.009727415,0.00007576076,0.003062562],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01017339,"threshold_uncertainty_score":0.02022833,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4995261141940527,"score_gpt":0.5734256507261446,"score_spread":0.07389953653209186,"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."}}