{"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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.01634939,0.00008602744,0.0002005809,0.000306763,0.001960876,0.00005622707,0.001816816,0.00004588232,0.00003810816],"category_scores_gemma":[0.007570075,0.00005925729,0.00001664904,0.000373848,0.0002080278,0.001672161,0.0009042946,0.000538619,0.000002356386],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003394641,"about_ca_system_score_gemma":0.001186407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00351187,"about_ca_topic_score_gemma":0.0007300028,"domain_scores_codex":[0.9968143,0.0005305169,0.001250107,0.0002746475,0.0008029462,0.0003275289],"domain_scores_gemma":[0.9960895,0.0008454992,0.00150794,0.000475563,0.0009512117,0.000130305],"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":[0.001116484,0.0002613643,0.3909715,0.0006635262,0.0000708845,0.000005029543,0.01669332,0.002210864,0.01051462,0.01997153,0.002057256,0.5554636],"study_design_scores_gemma":[0.0005393275,0.0001131757,0.02893343,0.0004127834,0.000005400209,0.00002000855,0.001176941,0.9641427,0.00004083134,0.000261372,0.004290948,0.00006302322],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2893102,0.00009953958,0.6969898,0.005521304,0.006555876,0.0008862792,0.0005171261,0.00002548044,0.00009439647],"genre_scores_gemma":[0.9220666,0.00004597924,0.07578558,0.0009796518,0.0004723773,0.00000400224,0.0006270153,0.000006942591,0.00001186872],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9619319,"threshold_uncertainty_score":0.9993384,"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."}}