{"id":"W4316174755","doi":"10.1177/14604582221115667","title":"Using ICD-9 diagnostic codes for external validation of topic models derived from primary care electronic medical record clinical text data","year":2023,"lang":"en","type":"article","venue":"Health Informatics Journal","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"North York General Hospital; University of Toronto","funders":"Canadian Institutes of Health Research","keywords":"Computer science; Diagnosis code; Topic model; Artificial intelligence; Face validity; Leverage (statistics); Natural language processing; Medicine; Population","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.007021947,0.00009150408,0.0003778439,0.0001497168,0.0006006038,0.00008820649,0.000544369,0.000111876,0.0000861754],"category_scores_gemma":[0.002618672,0.00008256667,0.0001221152,0.0003137609,0.0001034274,0.000585545,0.0001344766,0.0003559609,0.000006617039],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000259272,"about_ca_system_score_gemma":0.003964704,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005458252,"about_ca_topic_score_gemma":0.0003916544,"domain_scores_codex":[0.996439,0.0005778766,0.001587232,0.0001093437,0.0008873998,0.0003991352],"domain_scores_gemma":[0.9946588,0.003600397,0.0009209695,0.0001908494,0.0003314509,0.0002974773],"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":[0.00006504024,0.00007518691,0.01440325,0.0003425591,0.0001535769,0.000003543607,0.02162595,0.01175585,0.000003630306,0.003284482,0.001155832,0.9471311],"study_design_scores_gemma":[0.001437008,0.0002256734,0.012027,0.0006351044,0.000135902,0.00001315051,0.008306137,0.8813348,0.00001031996,0.08614843,0.009476694,0.0002497296],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3276977,0.0004905903,0.6700138,0.0008293266,0.0004911738,0.0002155627,0.00005887527,0.00002406081,0.0001789447],"genre_scores_gemma":[0.631803,0.01032333,0.353547,0.001454448,0.002050013,0.00001058979,0.0007515593,0.00002102551,0.00003903463],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9468814,"threshold_uncertainty_score":0.7033215,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2999121611428252,"score_gpt":0.5167997459878955,"score_spread":0.2168875848450703,"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."}}