{"id":"W2891010287","doi":"10.23889/ijpds.v3i4.946","title":"Improving the Coding Completeness of Hypertension in Inpatient Administrative Health Data Using Machine Learning Methods","year":2018,"lang":"en","type":"article","venue":"International Journal for Population Data Science","topic":"Medical Coding and Health Information","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Diagnosis code; Coding (social sciences); Logistic regression; Missing data; Computer science; Chart; Medicine; Random forest; Statistics; Machine learning; Data mining; Artificial intelligence; Population; Mathematics; Environmental health","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.0280744,0.0008971145,0.001022602,0.006484563,0.0008081783,0.00258519,0.00183984,0.0009264881,0.001391506],"category_scores_gemma":[0.1104388,0.0004598075,0.00125255,0.003964825,0.000691389,0.002226767,0.00217898,0.001537202,0.0005635754],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002230202,"about_ca_system_score_gemma":0.003764453,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02145079,"about_ca_topic_score_gemma":0.01286143,"domain_scores_codex":[0.9794272,0.0127423,0.001808287,0.00246676,0.002863378,0.0006921146],"domain_scores_gemma":[0.8739274,0.08697041,0.01173321,0.008535647,0.01776003,0.001073333],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003196096,0.0004404598,0.5890777,0.0004563181,0.0005151273,0.00009376751,0.0006284435,0.06972837,0.001093585,0.002668631,0.009756707,0.3252212],"study_design_scores_gemma":[0.00004447669,0.0001976053,0.1177927,0.0003872076,0.0001160307,0.00009481485,0.0005143285,0.864921,0.002876447,0.009646382,0.003329551,0.00007960247],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5964128,0.001897668,0.3843621,0.00333068,0.0002441501,0.0005794254,0.007681819,0.001781061,0.003710327],"genre_scores_gemma":[0.8013371,0.0004408661,0.1842552,0.0004374903,0.0002604953,0.0003455439,0.01208716,0.00009099,0.0007451526],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9719256,"threshold_uncertainty_score":0.1484734,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.7960701398576574,"score_gpt":0.6418011185677549,"score_spread":0.1542690212899025,"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."}}